Skip to main content

Full text of "Journal of the Society of Motion Picture and Television Engineers"

See other formats


From the collection of the 



m 

o Prelinger 
v Jjibrary 






San Francisco, California 
2007 




JOURNAL OF THE 
SOCIETY OF 

MOTION PICTURE 

AND 

TELEVISION 

ENGINEERS 




THIS ISSUE IN TWO PARTS 
Part I, June 1953 Journal Part II, Index to VoL 60 



VOLUME 60 
January June 1953 



SOCIETY OF MOTION PJ& T U R E 
AND TELEVISION ENGINEER'S 

40 West 40th St., New York 18 



CONTENTS Journal 

Society of Motion Picture and Television Engineers 

Volume 60 : January June 1953 



Listed below are only the papers and major reports from the six issues. See the 
Volume Index for those items which generally appear on the last few pages of each 
issue: Standards, Society announcements (awards, Board meetings, committee 
reports, conventions, engineering activities, membership, nominations, section 
activities), book reviews, current literature, letters to the Editor, new products and 
obituaries. 



January 

Frlm Projection Using Image-Orthicon Cameras . . . R. D. CHIPP 1 

Shooting Live Television Shows on Film KARL FREUND 9 

Practical Aspects of Reciprocity-Law Failure J. L. TUPPER 20 

A Method of Lighting Large Fields for High-Speed Motion Picture 

Photography HARRY R. CLASON 30 

X-Ray Motion Picture Camera and Printer for 70mm Film .... 

S. A. WEINBERG, J. S. WATSON and G. H. RAMSEY 31 
Application of Wide-Angle Optics to Moderately High-Speed Motion 

Picture Cameras H. E. BAUER and A. W. BLAKE 38 

New Automatic Film-Threading Motion Picture Camera 

G. J. BADGLEY and W. R. FRASER 49 

Animation Stand of New Design E. H. BOWLDS 58 

February 

Rapid Drying of Normally Processed Black-and- White Motion Picture 

Film F. DANA MILLER 85 

Further Experiments in High-Speed Processing Using Turbulent Fluids 

LEONHARD KATZ and WILLIAM F. ESTHIMER 105 

Isotransport Camera for 100,000 Frames Per Second 

C. DAVID MILLER and ARTHUR SCHARF 130 
Photographic Instrumentation in the Study of Explosive Reactions . 

MORTON SULTANOFF 145 

Television Camera Equipment of Advanced Design L. L. POURGIAU 166 
Splicing Motion Picture Safety Film Without Cements or Adhesives . 

LEONARD A. HERZIG 181 

ii Contents: Journal of the SMPTE Vol. 60 



March 

Sound-on-Film Recording Using Electrooptic Crystal Techniques . . 

ROBERT DRESSLER and ALBERT A. CHESNES 205 
An Intermediate Positive-Internegative System for Color Motion 

Picture Photography 

C. R. ANDERSON, N. H. GROET, C. A. HORTON and D. M. ZWICK 217 

Kinescope Recording Film Exposure Control 

RALPH E. LOVELL and ROBERT M. ERASER 226 
Time-Zone Delay of Television Programs by Kinescope Recording . 

RALPH E. LOVELL 235 
History and Present Position of High-Speed Photography in Great 

Britain W. DERYCK CHESTERMAN 240 

Rapid-Sequence Camera Using 70mm Film . CHARLES A. HULCHER 247 
Precision Film Editor Utilizing Nonintermittent Projection ..... ... 

TORBEN JOHNKE 253 

The Bridgamatic Developing Machine ,. 

JOSEPH A. TANNEY and EDWARD B. KRAUSE 260 
The Stereoscopic Art (A Reprint) JOHN A. NORLING 268 

April Part I 

Recommendations of the National Television System Committee for 

a Color Television Signal A. V. LOUGHREN 321 

Eidophor System of Theater Television .... EARL I. SPONABLE 337 
The Fischer Large-Screen Projection System (A Reprint) . E. BAUMANN 344 
Review of Work on Dichroic Mirrors and Their Light-Dividing Char- 
acteristics MARY ELLEN WIDDOP 357 

Television Recording Abstract W. D. KEMP 367 

Synchro-Lite Powered 16mm Projector ............. ! 

R. E. PUTMAN and E. H. LEDERER 385 

New Professional Television Projector W. E. STEWART 390 

High-Speed Photographic Techniques for the Study of the Welding 

Arc I. L. STERN and JOHN H. FOSTER 400 

Use of Photography in the Underground Explosion Test Program, 
1951-1952 R. M. BLUNT 405 

April Part II 

Manufacture of Magnetic Recording Materials 

EDWARD SCHMIDT and ERNEST W. FRANCK 453 
Commercial Experiences With Magna-Stripe . . EDWARD SCHMIDT 463 

Magnetic Striping Techniques and Characteristics 

B. L. KASPIN, A. ROBERTS, JR., H. ROBBINS and R. L. POWERS 470 
Magnetic Striping of Photographic Film by the Laminating Process . 485 

A. H. PERSOON 



Contents: Journal of the SMPTE Vol. 60 



iii 



Magnetic Sound Tracks for Processed 1 6mm Motion Picture Film 

THOMAS R. DEDELL 491 

Notes on Wear of Magnetic Heads 

G. A. DEL VALLE and L. W. FERBER 501 

A Study of Dropouts in Magnetic Film .... ERNEST W. FRANCK 507 
Methods of Measuring Surface Induction of Magnetic Tape .... 

J. D. BICK 516 
Standardization Needs for 16mm Magnetic Sound . E. W. D'ARCY 526 

May 

Progress Committee Report GEORGE R. GROVES 535 

Color and Reflectance of Human Flesh 

ALLEN STIMSON and EDWARD FEE 553 

Televising a Symphony Orchestra RUDY BRETZ 559 

Influence of Echoes on Television Transmission . . PIERRE MERTZ 572 
Erratum : Recommendations of the National Television System Com- 
mittee for a Color Television Signal 596 

Applications of High-Speed Photography in Rocket Motor Research . 

FLOYD G. STRATTON and KURT R. STERLING 597 
Simple Electronic Devices for High-Speed Photography and Cine- 
matography P. FAYOLLE and P. NASLIN 603 

June 

Resolution in Stereoscopic Projection . . . BERNARD G. SAUNDERS 651 
Depth Perception; With Special Reference to Motion Pictures A 

Reprint THADDEUS R. MURROUGHS 656 

70mm Test Vehicle Recorder CHARLES T. LAKIN 671 

High-Speed Motion-Picture Photography of Electrical Arcs on a 

High- Voltage Power System 

EVERETT J. HARRINGTON and HAROLD C. RAMBERG 675 
Addendum to Progress Committee Report : Developments in Germany 

GEORGE R. GROVES 680 

Visual Examination of 16mm Photographic Sound Tracks 

O. L. GOBLE 688 

Processing 16mm Color Film With a Silver Sound Track 

JOHN FRITZEN 690 

Matching Densitometry to Production .... HOWARD T. RAFFETY 692 
Transmission Densitometer for Color Films ... K. G. MACLEISH 696 
Motion-Picture Sound Installation at the University of California at 

Los Angeles BARRY EDDY 709 

Improved Equipment for Drive-in Theaters . . . . R. H. HEACOCK 716 
Drive-in Theater Dub'l Cone In-a-Car Speaker . J. ROBERT HOFF 721 

iv Contents: Journal of the SMPTE Vol. 60 



Film Projection Using 
Image-Orthicon Cameras 



By R. D. GHIPP 



Presented here are the results of over a year's use of image-orthicon cameras 
for all film transmitted by television station WABD, New York, totaling ap- 
proximately 2000 hr. In addition to brief consideration of the technical 
problems encountered, cost, reliability, convenience and other operational 
factors are discussed. 



o 



VER THE PAST two years there have 
been a number of discussions concerning 
the use of the image-orthicon pickup 
tubes for the transmission of film. 1 " 3 
These have covered, in some detail, 
the characteristics of such tubes, and 
the basic design of film projectors for 
television. They have also suggested 
some of the advantages and the dis- 
advantages of film cameras using image 
orthicons. The Research Division of 
the Allen B. Du Mont Laboratories, 
Inc. commenced tests of this type 
of pickup in 1948. By early 1950 
image-orthicon techniques were such 
that practical operating tests were in 
order. WABD in New York then 
installed one unit. The results led us 
to conclude that, for the broadcasting 
of "run-of-the-mill" available film, often 
without system preview or rehearsal, 
the image-orthicon camera had several 
desirable features. By early 1951, 
WABD was using Type 5820 Image 
Orthicons for all film transmission. It 



Presented on October 6, 1952, at the 
Society's Convention at Washington, D.G., 
by R. D. Ghipp, Director of Engineering, 
Du Mont Television Network, 515 Madison 
Ave., New York 22, N.Y. 



should be emphasized that we were 
primarily concerned with consistently 
good reproduction of films of uncertain 
vintage and quality, rather than with 
excellent reproduction of a few films 
especially made and processed foi 
television. 

Projectors 

The WABD projection room was 
originally laid out in March 1946, and 
equipped with two Simplex 35mm and 
one Victor 16mm projectors. These 
had been modified for 2-3-2-3 pull- 
down. In 1948 we added two Du Mont- 
Holmes, Model 5130G, 16mm television 
projectors. These were placed and 
mounted as shown in Fig. 1 . Mounting 
details for the 16mm projectors, which 
weigh approximately 300 Ib, are shown 
in Fig. 2. The concrete base weighs 
approximately 500 Ib and provides 
extremely steady operation. Tests for 
picture stability are better than the 
proposed RTMA/SMPTE specifica- 
tions, and no mechanical changes have 
been necessary to adapt any of the 
projectors to image-orthicon use. 

Cameras 

The cameras are standard Du Mont 
equipment, Model TA-124, normally 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 1. WABD projection room. 




Fig. 2. Mounting details of 16mm 
projector. 

used in studios and in the field. Each is 
equipped with a 90-mm //3.5 lens, and 
so placed with respect to the screen 
that the photo cathode is 1 8f in. distant. 
The only change required was reversal 
of the horizontal sweep, which is ac- 
complished by switching two leads on 



the deflection yoke. The cameras are 
aligned and adjusted in a conventional 
manner, using the "knee-of-the-curve" 
technique. 

Screen 

The original projection room was 
equipped with iconoscope cameras 
mounted on tracks. These were moved 
into position in front of any one of the 
four available projectors, which were 
separated from the cameras by a fire 
wall. This was a conventional arrange- 
ment. In order to substitute image- 
orthicon cameras, with no disruption of 
a heavy operating schedule, it was 
decided to retain the same method of 
camera mounting. This, in turn, pre- 
cluded use of direct projection, and 
indicated use of an intermediate screen 
between the projector and camera. 
The 16mm projectors, originally 



January 1953 Journal of the SMPTE Vol. 60 



equipped with 4-in. lenses for iconoscope 
use, were refitted with 2-in. //1. 9 lenses 
to produce a 3f in. X 4j in. image on 
the screen. The 35mm projectors, origi- 
nally equipped with 8^-in. lenses, were 
refitted with 5-in. //1. 9 lenses. Tests of 
many screen materials were made. 
Among the materials tested were tracing 
paper, standard rear-projection ma- 
terial, experimental translucent plastic, 
latex, ground glass and flashed opal. 
In view of the relatively small image 
size, most of these materials were dis- 
carded because of excessive grain. From 
the standpoint of minimum grain and 
minimum light dispersion, latex ap- 
peared to be superior to other materials. 
However, it aged rapidly, changed 
color, and was difficult to keep clean. 
Ground glass, which did not have these 
undesirable characteristics, was finally 
selected as satisfactory for practical use. 
A metal hood is used to prevent stray 
light from reaching the screen from 
either side. Figure 3 shows the hood 
from the projection-room side, and 
Fig. 4 from the camera side. Note in 
Fig. 4 the detent mechanism which 
permits the rapid movement and precise 
location of the cameras. 

Light Reduction 

As is well known, substitution of 
image orthicons for iconoscopes requires 
a substantial reduction of the light 
intensity to secure operation of the 
pickup tube at the proper point on the 
characteristic curve. Many means of 
accomplishing this have been suggested. 
We elected an extremely simple method : 
the substitution of a 300-w projection 
lamp for the usual 1000-w lamp. These 
lamps are operated at 90 v instead of the 
nominal 1 1 5 v. Screen brightness meas- 
urements, with no film in the gate, 
showed 125 ft-L, uniform within ap- 
proximately 10 ft-L. In order to 
reduce the light output of the 35mm 
projectors to equal that of the 16mm 
projectors, we dropped the arc current 



from 25 amp to 20 amp and added 
neutral density filters having 40% 
transmission. With the opening of the 
Du Mont Tele-Center in New York, 
new and less expensive light sources, 
now under investigation, will be used. 

Operation and Adjustment 

Image-orthicon cameras are equipped 
with vertical and horizontal saw-tooth 
controls for shading. We have found 
that these can be set for a particular 
pickup tube and specific light input, 
and no shading adjustments need be 
made during the running of a film. 
Figure 5 shows the operating position 
at WABD. One operator handles two 
cameras, plus remote control of a flying 
spot scanner and automatic slide 
changer. The Du Mont cameras in- 
corporate a black peak clipper as well 
as a white clipper. The white clipper 
is set so that normal white is never 
saturated, but extreme highlights may 
be reduced in amplitude. The black 
peak clipper is set so as to maintain 
constant black level and thus maintain 
standard setup. With most film, and 
a Type 5820 Image Orthicon, the 
camera lens may be stopped down to 
//5.6. When using a 5826, a typical 
aperture is //3.5. These lens settings 
are average. We have observed that 
the sensitivity of image orthicons of 
equal age may vary from tube to tube 
by a factor of 5. Sensitivity may also 
change with age by a factor of as much 
as 10. These variations are equalized 
by changing the lens stops. As indicated 
above, we have used both 5820 and 5826 
for film transmission. The 5826 pro- 
vides improved signal-to-noise ratio, 
and a somewhat better gray scale. 
Under some circumstances, with very 
dark scenes or very dense film, some 
adjustment to the iris is made. Also 
on occasion video gain and target 
voltage may be varied in order to avoid 
excessive black saturation. However, 



R. D. Chipp: Image-Orthicon Film Pickup 




Fig. 3. Light shield in front of 35mm projector. 




Fig. 4. Light shield on camera side of fire wall. 
January 1953 Journal of the SMPTE VoL 60 




Fig. 5. Film camera operating position. 



we have found that very little "gain- 
riding"* is necessary on average film. 

Cost 

There has been considerable discussion 
of the cost of using image orthicon tubes 
for film projection. Table I shows the 
life records of iconoscopes used in 1948, 
1949 and part of 1950, together with 
those 5820 Image Orthicons used for 
film from mid 1950 to early 1952. 
Experience to date indicates that 5826 
and 5820 Image Orthicons have com- 
parable life when used for film. The 
average hourly costs, based on list 

* "Gain-riding," an expression which 
originated in sound broadcasting, origin- 
ally referred to the frequent and sometimes 
continuous adjustment of audio amplifier 
gain controls to compensate for changes in 
program volume. The term has carried 
over into television broadcasting and may 
refer, in addition to its original meaning, 
to adjustment of the various video controls 
to compensate for changes in picture 
content. 



prices, are 25.7j and 83.8^, respectively. 
Table I shows also the hourly cost of 
projection lamps, which is believed in 
this instance to be of significance. 

The iconoscopes were generally re- 
moved from service for loss of resolution 

Table I 





Image 
orthicon 
(5820) 


Icono- 
scope 
(1850A) 


Hours 

Average hours 

Cost/hour 
Proi. lamp cost /he 


2200 
2660 
765 
665 
1407 
666 
804 
2285 

1431.5 


1256 
1481 
1533 
1441 
2660 
1876 
3590 
3003 

2102.5' 


$0.838 
mr 0.128 


$0.257 
0.660 



Total cost /hour $0 . 966 $0.917 



R. D. Chipp: Image-Orthicon Film Pickup 




Fig. 6A. A frame reproduced on an iconoscope system, 




Fig. 6B. A frame reproduced on an iconoscope system. 
January 1953 Journal of the SMPTE Vol. 60 




Fig. 7 A. The same frame as in Fig. 6A reproduced on an image-orthicon system. 




Fig. 7B. The same frame as in Fig. 6B reproduced on an image-orthicon system. 
R. D. Chipp: Image-Orthicon Film Pickup 



and low output, whereas the orthicons 
are generally removed, as is the case in 
studios, for "sticking."* Note that when 
tube and lamp costs are added, the 
expense of using iconoscopes is 92 per 
hour and the expense of using image 
orthicons is 97^ per hour. 

Conclusion 

Admittedly, an iconoscope chain, 
carefully modified and maintained and 
skillfully operated, can produce ex- 
tremely good pictures from good film. 
However, it has been our experience at 
WABD that the image orthicon camera 
can also produce good pictures, with 
no operational difficulty, from nearly 
all grades of film. Moreover, to the 
broadcaster, there are certain other 
advantages that may be gained from the 
use of image orthicons. Technical man- 
hours used for system previews and film 
rehearsals may be eliminated. When all 
cameras in a station are of the same type, 
maintenance procedures may be stand- 
ardized and simplified. Further, spare 
parts and tube inventories may be 
reduced. Finally, the operating cost is 
not appreciably in excess of that of the 
iconoscope. Figures 6A and 6B are 
frames from film reproduced on an 
iconoscope system, and teletranscribed 
directly from the line. Figures 7A and 
7B are the same frames from the same 
film, as reproduced on an image orthicon 
system and teletranscribed on the same 
facilities. Although many conclusions 
may be drawn from careful analysis of 
these pictures, it may be said that the 
second picture does not suffer by 
comparison. 

References 

1. R. L. Carman and R. W. Lee, "Image 
tubes and techniques in television film 

* "Sticking" refers to a phenomenon which 
occurs in image orthicon tubes, wherein, 
when a camera is panned from one scene 
to another, the first scene is briefly retained 
on the photocathode. As tubes age, the 
period of retention increases, and the tube 
is finally no longer satisfactory for broad- 
cast use. 



camera chains," Jour. SMPTE, 56: 
52-64, Jan. 1951. 

2. K. B. Benson and A. Ettlinger, "Prac- 
tical use of iconoscope and image 
orthicons as film pickup devices," Jour. 
SMPTE, 57: 9-14, July 1951. 

3. P. J. Herbst, "Televised film," Broad- 
cast News, May- June 1952. 

Discussion 

George Lewin (Signal Corps Photo Center): 
If there's a problem due to grain of the 
screen material, can't that be eliminated 
by just using a larger image on the screen? 

Mr. Chipp: Yes, it could. Perhaps I 
didn't stress sufficiently the physical prob- 
lem that we had in using an existing loca- 
tion. Now, when we move to our new 
studios if we do not use direct projection 
into the tube, we will use a larger screen. 

Anon: I wonder if you must form a 
first image on the screen. I'm not sure 
what your optical system is, but it would 
seem that the screen can be eliminated. 

Mr. Chipp: Well, the optical system is 
conventional in both the projector and the 
camera, so that we form an image on the 
screen. . . . 

Anon: There is then no reason really 
to form a first image. All you need is a 
field lens between two lenses. 

Mr. Chipp: Yes, that's probably true. 
I think investigation might show the field 
lens would be something rather expensive. 

Anon: Also, a field lens could be very 
cheap and very simple as long as it's in 
the plane of the first image. Every micro- 
scope uses this in almost every optical sys- 
tem. It's used to carry an image through 
several lenses. 

Mr. Chipp: That might be the case. 
We haven't fully investigated that. 

Barton Kreuzer (RCA, Camden, N.J.): 
What do you do in case of stills? 

Mr. Chipp: We use the flying-spot scan- 
ner for all slides. You didn't see the scan- 
ner in the picture, but the controls for the 
scanner are next to the video operator 
below the orthicon camera controls. In 
the case of a program where there may be 
a series of titles, we use title cards in the 
studio. They're picked up by the studio 
image orthicon. 

Mr. Kreuzer: In these iconoscope com- 
parisons, did the iconoscope chain have 
the latest improvements in it that were 
described to the industry and pretty 
generally accepted about six or eight 
months ago? 

Mr. Chipp: Yes, to the best of my knowl- 
edge. 



January 1953 Journal of the SMPTE Vol. 60 



Shooting Live Television 
Shows on Film 



By KARL FREUND 



Experience in shooting live television shows on film is described, in which 
three motion picture cameras were used instead of television cameras, with 
overhead lighting and in the presence of an audience. Subject contrast was 
measured by means of a flare-free brightness photometer. 



JL HERE ARE various methods by 
which programs are produced for home 
television: (1) the direct live show with 
audience participation, (2) the same 
show kinescope recorded, (3) motion 
picture films formerly made for theater 
exhibition and (4) motion pictures 
especially made for television by the use 
of one camera and lights properly 
placed for each individual camera setup. 
In 1951 Desilu Productions (Desi 
Arnaz and Lucille Ball) asked me to be 
Director of Photography for the / Love 
Lucy audience-participation show, intro- 
ducing for the first time a deviation from 
standard procedure. Television cam- 
eras were to be replaced by three motion 
picture cameras to provide more flexi- 
bility in editing and nation-wide better 
photographic quality than that accom- 
plished by kinescope recording. 



Presented on October 7, 1952, at the 
Society's Convention at Washington, D.C., 
by John W. Boyle for the author, Karl 
Freund, Director of Photography, 15024 
Devonshire St., San Fernando, Calif. 



Being aware that this was a step in the 
right direction and a challenge to a 
motion picture camerman, I accepted 
the assignment without realizing that be- 
sides the usual problems connected with 
photographing motion pictures, I would 
inherit additional troubles photograph- 
ing a live television show. 

A regular motion picture studio was 
equipped with bleachers to accommodate 
approximately 300 people and a series of 
directional microphones and loudspeak- 
ers installed overhead (Fig. 1). The 
lighting for the sets had to be placed 
above, to give the audience a clear view 
of the show and also to give the cameras 
100% mobility without interference of 
floor cables. 

Motion picture technicians have ex- 
pressed a special curiosity as to why it 
was decided to present the / Love Lucy 
show before a live audience. It seemed 
unusual to make so many painstaking 
preparations to please a group of only 
300 spectators each week when the show 
was aimed at an ultimate audience of 
many millions. And yet the one thing 
which may be the key to the popularity 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 1. Regular motion picture studio equipped with bleachers to accommodate 

approximately 300, with directional microphones 

and loudspeakers installed overhead. 



10 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 2. Lights placed above the set to give the audience clear 
view of show and to give the cameras 100% mobility. 



of the program is the long accepted fact 
that an audience has an astonishing 
effect in stimulating performers. It also 
has been generally acknowledged that 
laughs dubbed in later sound highly 
artificial. 

It naturally would have been com- 
paratively more simple to produce / Love 
Lucy in routine fashion, and certainly a 
better guarantee of good photographic 
and composition results would have been 
accomplished through ordinary methods 
of setups and close-ups with as many re- 
takes as necessary. 

Since it was necessary to place the 
lights overhead (Fig. 2) and at the same 
time do photographic justice to the 
actors, the natural thing was for me to 
begin with a series of tests. After pro- 
jecting these tests in the laboratory pro- 
jection room, I found they had the qual- 



ity I was accustomed to when photo- 
graphing motion picture comedies, but 
when viewed over a closed television 
circuit, these prints showed too much 
contrast. 

Seeking an explanation for this, I was 
briefed by television engineers and in- 
formed that the iconoscope tube, which 
is most used for monochrome film tele- 
casting, has certain limitations not yet 
overcome by the manufacturer and I 
was cautioned to keep this in mind and 
reduce my lighting contrast and sup- 
press the brightness range considerably. 
Next, I familiarized myself with current 
television publications of film manufac- 
turers from which I quote:* 

*"The Use of Motion Picture Films in 
Television," Eastman Kodak Co., 1949, 
p. 12. 



Karl Freund: Shooting Television Films 



11 




Fig. 3. Brightness spot meter to 
measure brightness of a very small area 
at any distance from 4 ft to infinity. 

"The most notable departure from 
standard motion picture technique in 
making films for television use is that 
relating to the subject lighting contrast 
which is required. The limited range 
of brightnesses which can be reproduced 
as satisfactory tone scale values in the 
television system imposes restrictions on 
the range of brightnesses which can be 
effectively reproduced on a receiver 
screen from a subject being televised." 

It should be noted that the term 
"lighting contrast" does not mean sub- 
ject contrast or subject brightness range. 

The true subject contrast or subject 
range is usually much higher than the 
lighting contrast since it takes into 
account the different reflectances of the 
various elements of the scene. It can be 
measured accurately only by means of a 
flare-free telescopic type of brightness 
photometer (Fig. 3) which measures an 
extremely small area and which allows 
the instrument to be situated at a suffi- 
cient distance so as not to obstruct any 
light falling on the subject. In film tele- 
casting the subject is an image on film, 



which means that the density range must 
not exceed a certain value if good tone 
reproduction is to be obtained both in 
highlights and shadows. 

I should quote further : 

"The question which immediately 
arises is what method to use in order to 
obtain the desired density compression 
in the positive print. Upon first ex- 
amination it might appear that this 
might be accomplished equally well in at 
least three different ways : 

(1) In exposing the original negative, 
use a subject lighting contrast which is 
considerably lower than that which is 
normally used for conventional black-and- 
white motion picture photography, and 
process both the negative and print in 
the normal way. 

"(2) Use normal lighting contrast and 
exposure but alter the processing condi- 
tions of negative or positive or both, to 
obtain an overall reproduction gamma 
which is lower than normal. 

"(3) Use normal lighting contrast and 
exposure, process the negative and posi- 
tive in the usual manner, but make the 
print 2 or 3 or more printer steps lighter 
than what would be desirable if the print 
were to be used for normal projection 
purposes."* 

In shooting / Love Lucy I selected the 
first method since this involved no de-, 
partures from standard practice in proc- 
essing laboratory operations. 

One important point I want to men- 
tion at this time is that in viewing the 
first show over my own television re- 
ceiver, I realized that I do not have com- 
plete control of the end results. There is 
an engineer in every control booth when 
the show is televised who can change the 
screen image according to instructions 
and depending upon the condition of the 
equipment he has to work with; besides, 
there are millions of individual television 
owners who also have control over the 
quality of the final product on their own 
screens. Disturbed by all these condi- 



Ibid., p. 13. 



12 



January 1953 Journal of the SMPTE Vol.60 



REHEARSAL SCHEDULE FOR "I LOVE LUCY" #46 
Friday, September 5, 1952 



8:00 A.M. 



9:00 A.M. 
10:30 A.M. 
11:00 A.M. 

11:30 A.M. 
11:45 A.M. 

12:00 Noon 

to 
12:30 P.M. 

12:00 Noon 

to 
3:00 P.M. 

3:00 P.M. 

to 
4:00 P.M. 

4:00 P.M. 

to 
7:00 P.M. 

7:00 P.M. 

to 
8:00 P.M. 



ELECTRICAL 

Gaffer 7 " 

Best Boy 

3 Operators 

2 Dimmer Men 

1 Gen. Oper. (GSS) 

1 Stand-By Laborer 



GRIPS 

Noble Craig 
2nd Grip 



Karl Freund 
Jack Owen 

2 Cable Men 

2 Stand-ins 

3 Asst . Cameramen 
1 Mike Man 

3 Dolly Grips 

3 Camera Operators 

Lunch for early crew 



REHEARSAL WITH CAMERAS 



Lunch for late crew and cast 



REHEARSAL WITH CAMERAS 



DRESS REHEARSAL CAST 



N. Craig 
F. Jenkins 
Best Boy 
Jack Owens 



Fig. 4A. Typical schedule for first day rehearsal with cameras. 
Karl Freund: Shooting Television Films 



13 



SHOOTING SCHEDULE FOR "I LOVE LUCY" #46 
Saturday, September 6, 1952 

12:30 P.M. ELECTRICAL GRIPS 



Gaffer Head Grip 

Best Boy Floor Grip 
2 Operators 

1 Floor Man 

2 Dimmer Men 

1 Gen. Oper. (GSS) 

12:30 P.M. Hairdresser B. French 

1:00 P.M. 3 Asst. Cameramen 

SOUND 



Mixer 1:30 P.M. 
Mike Man 1:00 P.M. 
3 Cable Men 1:00 P.M. 

3 Dolly Grips 
1:00 P.M. 



2:00 P.M. 


Karl Freund 


2:00 P.M. 


REHEARSAL WITH CAST 


to 




5:30 P.M. 




4:00 P.M. 


1 Camera Loader-Recorder 




1 Makeup Man Hal King 


5:30 P.M. 


Dinner for cast and crew 


to 




6:30 P.M. 




6:15 P.M. 


MUSIC BALANCE and RECORDING 




from stage 


6:30 P.M. 


TALK THRU Conference with Bill 


to 


Asher (all crew except sound 


7:15 P.M. 


man) 


7:15 P.M. 


Doors Operu 


7:45 P.M. 


Warm-up 


8:00 P.M. 


SHOW 


to 




9:30 P.M. 




9:45 P.M. 


Start pickup shots 



Fig. 4B. Typical schedule for second day rehearsal with cameras. 
14 January 1953 Journal of the SMPTE Vol. 60 




Fig. 5. The only floor lights used are mounted on the bottom of each camera dolly; 
movements of the dollies are marked on floor for each individual scene. 



tions, I was advised by television 
authorities to be. patient, compromise 
with the young industry and trust in 
future developments which in time 
would give us the quality we are accus- 
tomed to seeing in motion pictures. 

The following is an outline of the prac- 
tice established in shooting the / Love 
Lucy and Our Miss Brooks shows. 

Four days are required to line up each 
weekly show two days of which are 
taken by the director and the cast for re- 
hearsals. At the end of the second day 
I have an opportunity of seeing a run- 
through which enables me to make notes 
and sketches of positions to be covered 
by the cameras and to instruct the elec- 
trical crew as to where the lights are to be 
placed. The last two days are occupied 
by rehearsals with Cameras. Figures 4A 



and 4B show typical schedules for the 
two days of rehearsal with cameras. 
These schedules have to be kept up to the 
minute by everybody concerned, in- 
cluding the cast, because a show with 
audience participation must go on at a 
specified time. The actual shooting 
time for the entire show is approximately 
one and a half hours. 

The cameras used are three BNC 
Mitchells with T-stop calibrated lenses 
on dollies. The middle camera usually 
covers the long shot using 28-mm to 50- 
mm lenses and the two close-up cameras 
75 to 90 apart from the center camera 
using 3- or 4-in. lenses, depending upon 
certain requirements for coverage. 

Mounted on the bottom of each cam- 
era dolly and above the lens, controlled 
by dimmers, are the only floor lights used 



Karl Freund: Shooting Television Films 



15 




Fig. 6. Electricians handling the dimmer board. 




Fig. 7. Overhead lights using 6 1-K silver-coated lamp bulbs for indirect lighting. 
16 January 1953 Journal of the SMPTE VoL 60 



(Fig. 5). A crew of four men the 
operative cameraman, assistant, grip 
and cable man is required on each 
camera, and coordination among them is 
essential, unlike in television where the 
responsibility lies with only one man 
handling his own camera movements, 
focusing, tilting and panning and having 
the additional advantage of viewing his 
results immediately. 

Each movement of the dollies is 
marked on the floor for each individual 
scene (Fig. 5). The entire crew uses an 
intercommunication system, since all the 
movements of the camera are cued from 
the monitor box. I personally use a 
two-circuit intercommunicator to enable 
me to talk separately to the monitor 
booth and the camera crew on one, and 
to the electricians handling the dimmers 
and the switchboard on the other (Fig. 
6). 

It is a noteworthy coincidence that 
just 30 years ago this month, I introduced 
the moving camera to the motion picture 
industry in the German picture Last 
Laugh. If I had known then what 
trouble I was storing up for myself, 
especially in the / Love Lucy show where 
the cameras are moving constantly, I 
certainly would have thought twice be- 
fore starting this innovation. 

Since the audience is lost at the end of 
each show, retakes are not desirable be- 
cause laughs would then need to be 
dubbed in. Retakes are therefore made 
in emergency cases only. Even better- 
lit close-ups, taken by a single camera 
with properly placed lights, had to be 
discarded. Cutting these so-called 
"glamour" close-ups into the picture 
proved unsatisfactory as it was found 
that they stood out like a sore thumb. 
Retakes were made when necessary, and 
the same lighting used as during the 
show except for minor unnoticeable 
changes. 

The illumination level of 250 foot- 
candles measured with the incident light 
meter and the lighting contrast of 2 to 1 



are maintained practically over the en- 
tire set. 

The lens T-stop is 4.5. The per- 
missible brightness range which is gov- 
erned by both the illumination and the 
reflecting power of the various parts of 
the scene should not exceed a 20 to 1 
ratio, so special attention is paid to 
makeup, dresses, props and color of the 
sets. The best reproduction of face tone 
quality I experience with makeup two 
to three shades darker than usually 
applied in motion pictures. 

The following materials were of East- 
man Kodak manufacture: Plus X, 
35mm, type 5231 Negative developed to 
gamma 0.68; print on Fine-Grain Re- 
lease Positive, 35mm, type 5302, gamma 
2.40; Fine-Grain Duplicating Positive, 
35mm, type 5365, gamma 1.40; Fine- 
Grain Dupe Negative, 16mm, type 7203, 
gamma 0.55; Fine-Grain Release Posi- 
tive, type 7302, 16mm, gamma 2.15. 

New overhead lights were developed 
with six 1000-watt silver-coated lamp 
bulbs for indirect lighting (Fig. 7). For 
front key-light, converted 5000-watt 
pans with sleeves to accommodate dif- 
fusing material are used (Fig. 8) ; other- 
wise there is no deviation from standard 
motion picture lighting equipment. 

All the lights are preset for each indi- 
vidual scene and changed accordingly 
by signals to the switchboard operator. 
Experiments are now under way to 
eliminate many individual spot lights by 
replacing them with sealed beam lights 
and better lighting equipment is in de- 
velopment to enable easier operation for 
this type of show. 

To have had the opportunity to play a 
part in the success of the / Love Lucy show 
which is now rated the No. 1 television 
show in the nation assures me that the 
efforts to overcome the handicaps have 
not been in vain, and the results accom- 
plished are comparable to motion picture 
photographic quality where comedy 
treatment of lighting is required. 

In conclusion, I want to give credit to 
the producer, cast and crew, Dr. Nor- 



Karl Freund: Shooting Television Films 



17 




Fig. 8. Front key-light, converted 5-K pans with sleeves 
to accommodate diffusing material. 



18 



January 1953 Journal of the SMPTE Vol. 60 



wood Simmons of Eastman Kodak, Mr. 
Herb Pangborn of CBS and the staff of 
Consolidated Laboratories. Through 
their lively cooperation I am able to 
achieve the best photographic results. 

Discussion (replies written by the 
Author) 

Robert Af. Fraser (National Broadcasting 
Co., New York): How much time is spent 
in editing? While production time is 
four days up to the time of shooting there 
must be considerable time, of course, 
spent after the shooting. 

Dr. Karl Freund: Four days are not re- 
quired for shooting; as mentioned, only 
two days are used for rehearsal with actors 
and two days for rehearsal with cameras. 
The actual shooting time is approximately 



one hour and fifteen minutes. This in- 
cludes reloading the cameras, makeup 
changes and costume changes of the cast. 
The editing time is approximately four 
weeks. 

Mr. Fraser: Is the sound recorded after 
the show, the musical bridges, for example, 
which are used evidently to cover up time 
lapses in the actual show? 

Dr. Freund: The sound is recorded 
magnetically and the musical bridges are 
recorded before the actual shooting of the 
show to save expenses for special orchestra 
sessions. 

R. T. Van Niman (RCA Victor Div., 
Camden, N.J.): How many weeks ahead 
do they shoot the shows before they are 
shown on television screens? 

Dr. Freund: Shows you see on a television 
screen were shot six weeks ahead. 



Karl Freund: Shooting Television Films 



19 



Practical Aspects of 
Reciprocity-Law Failure 



By J. L. TUPPER 



The occasional failure of sensitometric data to provide a reliable indication 
of the practical performance of photographic materials is usually attributable 
to the failure of the reciprocity law. The effect of reciprocity-law failure on 
the characteristic D-log curves of various films is shown graphically. The 
influence of developing time and of the temperature of the film on the effec- 
tiveness of exposure at various intensity levels is discussed. Certain generali- 
zations are made about the failure of the reciprocity law which may be helpful 
in reconciling differences between laboratory measurements and the results 
obtained in motion picture practice. 



w, 



ITH THE general acceptance of 
the methods of photographic sensitom- 
etry in the control of the uniformity 
of production of motion picture nega- 
tives, prints, duplicates and sound 
records, and in the analysis of new 
techniques and processes, there has been 
a growing concern about occasional 
failures of this tool to provide a reliable 
indication of the performance of photo- 
graphic materials in practice. Similarly, 
there is the problem in high-speed 
photography of reconciling sensitometric 
data obtained under standard conditions 
with the effective characteristics of the 
material realized at extremely short 
exposure times. There are many pos- 



Communication No. 1518 from Kodak 
Research Laboratories, a paper presented 
on April 23, 1952, at the Society's Con- 
vention at Chicago, by J. L. Tupper, 
Eastman Kodak Co., Kodak Park Works, 
Rochester 4, N.Y. 



sible causes of these discrepancies, but 
the one most frequently responsible is 
the failure of the reciprocity law. It 
is only through an understanding of the 
practical manifestations of this phe- 
nomenon that sensitometry can be uti- 
lized most effectively in the wide variety 
of applications to which it may be put. 

An assumption which is frequently 
made in the substitution of a sensito- 
metric test pattern composed of sys- 
tematically graduated exposures for an 
actual pictorial or sound record is that 
there will be a one-to-one correspondence 
between the densities of the developed 
images resulting from equal exposures, 
regardless of the mechanism used in 
impressing the exposures on the sensitive 
material, or of the intensity and time 
components of the exposures, provided 
their products (/ X are in all cases 
equal. However, the density obtained 
is not uniquely determined by the value 



20 




January 1953 Journal of the SMPTE Vol. 60 



of the exposure, but it usually depends 
upon the individual values of / and t. 
This failure of time and intensity to act 
reciprocally is a consequence of the 
dependence of the latent-image-forming 
process on the rate at which the exposure 
event takes place. 

Latent-image theory suggests a mech- 
anism which explains the normal de- 
crease in efficiency of latent-image 
formation when the rate at which energy 
is received exceeds or falls below a 
certain optimum value. The tendency 
for latent images formed at high in- 
tensities to be less readily developable 
than those formed at low intensities 
has been explained on the basis of 
differences in the spatial distribution of 
the latent-image nuclei in the silver 
halide crystal. From a practical point 
of view, however, it must be reported 
that few photographic materials are 
affected exactly alike by changes in the 
parameters of time and intensity in the 
exposure equation. Strictly, in dis- 
cussing reciprocity failure, each material 
should be treated as a special case. On 
the other hand, there are certain typical 
patterns which characterize many films 
used in motion picture photography, 
and it is those which will be used to 
illustrate the practical significance of 
this phenomenon. 

The conventional method of repre- 
senting the effect of reciprocity-law 
failure is a curve in which the logarithm 
of the exposure required to produce a 
particular density is plotted as a function 
of the logarithm of the intensity of the 
exposing light. Such a curve based 
on a density of 1.0 is shown in Fig. 1. 
The parallel straight lines inclined at 
45 are lines of constant exposure time. 
These are merely parts of the reference 
framework. Curve X is the reciprocity- 
failure curve. If the reciprocity law 
held, that is, if the photographic effect 
depended only on the product of time 
and intensity, this curve would appear as 
a straight horizontal line. It is seen 
however, that the curve is characterized 



by a minimum, which corresponds to 
the optimum intensity. The curve rises 
at intensity values above and below the 
optimum. The amount of the upward 
turn of the curve at the two ends of the 
graph is a measure of the amount of the 
reciprocity failure at low and high 
intensities. The magnitude of the low- 
intensity failure may be conveniently 
expressed in terms of the ratio of the 
exposure at some arbitrarily selected 
low value of intensity, such as thai 
indicated at A, to the exposure at the 
optimum intensity level indicated at B. 
Similarly, the magnitude of the high- 
intensity failure may be expressed as the 
ratio of the exposure at an arbitrarily 
selected high intensity, C, to the ex- 
posure at the optimum intensity, B. 
In this example, the low-intensity 
failure, in terms of the ratio, A:B, is 
5.3 and the high-intensity failure, ratio 
G :B, is 2.0. This pattern is characteris- 
tic of most negative materials; the 
optimum intensity is about 8 meter- 
candles, corresponding to an exposure 
time of 1/40 sec, and the low-intensity 
failure is greater than the high-intensity 
failure. Also note that at the very high 
intensities (100,000 meter-candles, up- 
ward), the reciprocity law appears to 
hold over a considerable range. The 
upper limit of this range has not yet 
been determined experimentally, al- 
though there is some evidence that it 
may extend well beyond 10 7 meter- 
candles. 

Some mention should be made of 
differences in reciprocity failure in 
various types of photographic materials. 
In general, the greatest differences occur 
in the low-intensity region. This is a 
consequence of designing emulsions to 
meet the particular requirements of the 
various systems in which they are to be 
used. A large amount of low-intensity 
reciprocity failure is desirable if the 
material is to be used, for example, in 
motion picture photography, where 
camera exposures are made at 1/25 
sec, or less, since the lower sensitivity 



J. L. Tupper: Reciprocity-Law Failure 





E 

- 0.0 



1.0 



I0 4 " !0 3 " I0 2 " 10" l" 10"'" IO' 2 " IO' 3 " IO' 4 " IO' 5 " IO" 6 " 




3.0 T.O 1.0 3.0 5.0 

Log I (me) 

Fig. 1. Reciprocity-law-failure curve based on a density of 1.0. 



7.0 



1.0 



2.0 



3.0 



r KK K>- 



2" 10-3" 10-4* 




T.O 



1.0 

Log I ( m c ) 



2.0 



3.0 



Fig. 2. Reciprocity-law-failure curves based on six levels of density. 
22 January 1953 Journal of the SMPTE Vol. 60 



to low levels of illumination provides 
an additional measure of safety in the 
darkroom. On the other hand, if the 
material is normally exposed to very 
low levels of illumination, as in astro- 
nomical photography, an emulsion is 
selected which shows very little low- 
intensity reciprocity failure. Thus, in 
different photographic materials, the 
ratio, A:B, may range from 1.0 to as 
much as 10. A much smaller range is 
found in the ratio, C : B, for the various 
materials, being of the order of 1.5 to 4. 
There is also an appreciable variation 
in the optimum intensity with different 
materials, the value ranging from about 
1/10 meter-candle to 10 meter-candles. 

If, instead of plotting the reciprocity- 
failure curve for a single selected value 
of density, a family of curves is plotted 
for several density values, considerably 
more information can be obtained about 
the effects of this phenomenon on the 
sensitometric characteristics of photo- 
graphic materials. Such a curve is 
shown in Fig. 2. In this figure, the 
logarithm of the exposure required to 
produce densities of 0.1, 0.25, 0.5, 0.75, 
1.0 and 1.25 is plotted as a function of 
the logarithm of intensity. For clarity, 
an expanded scale is used and only part 
of the intensity range shown in the 
previous figure is represented. It is 
seen that these curves are approximately 
parallel. The vertical displacement of 
the curves indicates the logarithm of 
the exposure ratio required to increase 
the density from the lower to the higher 
specified density value when the in- 
tensity of exposure is constant and the 
time of exposure is varied. The log- 
exposure interval between the intercepts 
of the 45 lines and the reciprocity 
curves indicates the logarithm of the 
factor by which exposure must be 
increased to increase the density from 
the lower to the higher indicated value 
when the time of exposure is constant 
and intensity is varied. By plotting the 
density values of the different curves as 
a function of the logarithm of the corre- 



Table I. Values of Log Exposure 
Taken From Curves in Fig. 2. 

Log/X / 



Density 



B 



0.10 3.68 3.69 3.92 3.70 

0.25 2.07 2.07 2.27 2.08 

0.50 2.49 2.48 2.65 2.56 

0.75 2.89 2.87 I. 01 2.98 

1.00 1.29 1.27 T.35 1.43 

1.25 1.69 1.68 T.71 1.83 

spending exposure, characteristic H&D 
curves can be obtained for any selected 
value of time and intensity. 

Since the vertical displacement of the 
curves is related to the change in ex- 
posure required to produce the indicated 
density differences at a constant in- 
tensity level, exposure time being the 
variable, the H&D curve derived from 
utilizing this vertical displacement is, 
therefore, a "time-scale" curve. For 
example, it is possible to derive the 
H&D curve which would be obtained 
by exposing on an Eastman Type lib 
(Time-Scale) Sensitometer. In this 
sensitometer, the intensity is one meter- 
candle, shown at A in Fig. 2. The 
logarithm of the exposures corresponding 
to the various values of density at this 
selected value of intensity are shown 
by the small circles. The actual log 
exposure values are given in column A 
of Table I. 

To obtain an intensity-scale D-log E 
curve from the reciprocity-failure 
diagram, it is necessary to select the 
log E values at the intercepts of the 45 
lines and the reciprocity curves at the 
various density levels. This procedure 
is followed because in an intensity-scale 
exposure the exposure time is constant 
and the intensity is variable, and in this 
diagram the 45 lines represent the loci 
of points of constant time. If it is de- 
sired to obtain the intensity-scale curve 
which results from an exposure time of 
0.01 sec, the exposures corresponding to 
the intercepts of the 45 line at the 



J. L. Tupper: Reciprocity-Law Failure 



23 




Fig. 3. D-log E 
curves as follows: 

O = Time-scale at 
1.0 me 

A = Intensity- 
scale at 0.01 sec 

X = Intensity- 
scale at 1.0 
sec 

D = Intensity- 
scale at 
0.0001 sec 



Log E (mcs) 



20 



1.0 




0.0 



10 



2.0 



Log E (mcs) 



Fig.*4. D-log E curves 
of Eastman Fine Grain 
Sound Recording 

Safety Film. 
Curve A = Intensity- 
scale at 10 ~ 2 sec 
Curve B = Intensity- 
scale at 10 ~ 5 sec 



2.0 




0.0 



1.0 



2.0 



Log E (mcs) 



Fig. 5. D-log E curves 
of Eastman Fine Grain 
Duplicating Negative 

Safety Film. 
Curve A = Intensity- 
scale at 10 ~ 2 sec 
Curve B = Intensity- 
scale at 10-5 sec 



24 



January 1953 Journal of the SMPTE Vol. 60 




E (mcs) 



0.01 -sec position are determined. These 
values are indicated by the small 
triangles. The actual log E values are 
given in column B of Table I. In a 
similar manner, density-log exposure 
data can be obtained for an intensity- 
scale exposure of 1.0 sec (small crosses) 
and 0.0001 sec (small squares). The 
numerical values are shown in columns 
G and D, respectively, of Table I. 

Using the data obtained in this 
manner from the reciprocity-failure 
curves, the D-log E curves shown in 
Fig. 3 were plotted. An examination 
of these curves shows that the sensito- 
metric characteristics of a particular 
photographic material are dependent, to 
a considerable extent, upon the intensity 
and time parameters of the exposure 
and upon the method of exposure modu- 
lation (time scale vs. intensity scale). 
There is no simple generalization that 
can be made about the way in which 
the sensitometric characteristics of differ- 
ent materials change as the time and 
intensity parameters are altered, al- 
though normally the gradient of the D- 
log E curve decreases as the intensity 
level increases. Thus, a D-log E curve 



Fig. 6. D-log E curves 
for Eastman Fine 
Grain Release Positive 

Safety Film. 
Curve A = Time-scale 

at 55 me 

Curve B = Intensity- 
scale at 0.01 sec 
Curve C = Intensity- 
scale at 0.001 sec 



obtained from a 1-sec exposure (in- 
tensity-scale) will have a higher gamma 
than one exposed at 0.0001 -sec (in- 
tensity-scale). Other changes occur 
in the shape of the characteristic curve, 
notably in the toe region. Generally, 
the length of the toe increases with 
increasing intensity. The difference 
between a time-scale curve and an 
intensity-scale curve is normally least 
when the median intensity of the intensity 
scale is equal to the intensity of the time- 
scale exposure. Obviously the curves 
shown in Fig. 3, while typical, precisely 
characterize only one specific type of 
photographic material. More or less 
change in sensitometric curve shape may 
be expected to parallel larger or smaller 
failures in the reciprocity law for other 
films. 

An interesting example of a rather 
large difference of this sort is shown in 
Fig. 4. These curves are two D-log E 
curves for Eastman Fine Grain Sound 
Recording Safety Film intended for use 
with variable-density sound-recording 
equipment. Curve A was obtained from 
an intensity-scale exposure, the exposure 
time being 0.01 sec. Curve B represents 



L. Tupper: Reciprocity-Law Failure 



25 



an intensity-scale exposure, the exposure 
time being about 10~ 5 sec, which is of 
the order of exposure times used in 
sound recording. A large difference in 
curve shape is apparent, particularly 
the much longer toe on curve B. It is 
this characteristic which is realized in 
actual sound-recording applications that 
makes this film such an excellent com- 
panion for Eastman Fine Grain Release 
Positive Safety Film. 

A smaller difference in D-log E curves 
is found for Eastman Fine Grain Dupli- 
cating Negative Safety Film, as seen 
in Fig. 5. Here a change from an 
intensity-scale exposure of 0.01 sec to 
10~ 5 sec produces noticeably less differ- 
ence in curve shape than was shown in 
Fig. 4. 

It has been reported that the change 
in the characteristic curve of Eastman 
Fine Grain Release Positive Safety Film 
when the printer speed is increased is a 
matter of special interest in motion 
picture laboratories. In Fig. 6 are 
shown three D-log E curves. Curve A 
was obtained with the Eastman Type 
lib Sensitometer. The gamma of this 
curve is 2.6. Curve B, which is an 
intensity-scale curve based on an ex- 
posure of 0.01 sec, is similar to curve A 
in gamma, but it has a somewhat longer 
toe portion. Curve C, an intensity- 
scale curve for an exposure of 0.001 sec, 
is lower in gamma (7 = 2.4), and the 
toe portion is longer than it is in either 
of the other two curves. This explains 
why it has been found necessary in 
practice to alter the lib "control 
gamma" in order to maintain constant 
screen contrast when a major change is 
made in the printer speed. Changing 
the "control gamma" is, of course, 
equivalent to specifying a change in 
developing time to compensate for the 
reduction in the gradient of the D-log E 
curve obtained at the shorter exposure 
times. Thus, in the example shown in 
Fig. 6, the development-time increase 
required to raise the gamma of curve C 
to 2.6 would raise the gamma of curve 



A (lib control gamma) to something 
over 2.7. 

In addition to the differences in the 
reciprocity-failure characteristics of vari- 
ous photographic emulsions, there are 
a number of factors which alter the 
reciprocity failure of any particular 
emulsion. Of these, those associated 
with development are of most practical 
interest. Because of the dependence 
of the distribution of the latent-image 
nuclei within the silver halide crystal 
upon the intensity level of the exposure, 
the apparent reciprocity-failure charac- 
teristics are affected by the choice of 
developer and the extent of develop- 
ment. By the proper choice of proc- 
essing it is possible to confine develop- 
ment either to the surface latent image 
or to the internal latent image. With 
normal commercial developers, the sur- 
face latent image is primarily responsible 
for the initiation of development, but 
with prolonged development times the 
internal latent images may also con- 
tribute to the initiation of development. 
In Fig. 7 is shown a family of reciprocity 
curves resulting from five different 
development times in a commercial 
developer. The curves are based on the 
exposure required to produce a density 
of 0.20. It is seen that as development 
is increased, the high-intensity reciprocity 
failure diminishes so that, at very long 
development times, the curve is nearly 
horizontal at the high-intensity levels. 
Increasing development time has rela- 
tively little effect on the low-intensity 
reciprocity failure. This effect is ex- 
plained by the hypothesis that a rela- 
tively greater percentage of internal 
latent images are formed at high in- 
tensities than at low intensities and that 
with prolonged development the internal 
images become effective in the initiation 
of development. This suggests the de- 
sirability of fully developing films which 
are used in high-speed photography 
when maximum sensitivity is required. 

Of more than academic interest is 
the effect of the temperature of the film 



26 



January 1953 Journal of the SMPTE Vol. 60 



0.0 



~ 10 

8 1 



20 



3.0 




3.0 2.0 T.O 0.0 1.0 2.0 3.0 

Log I (m c ) 

Fig. 7. Reciprocity-law-failure curves for five developing times. 



1.0 



E 

r o.o 



i.o 




To" 



To" 



2.0 1.0 

Log I (me) 



00 



10 



Fig. 8. Reciprocity-law-failure curves showing effect of temperature during exposure. 
J. L. Tupper: Reciprocity-Law Failure 



2.0 



1.0 




2.0 1.0 

Log E (mcs) 



Fig. 9. D-log E curves 
of Eastman Plus-X 
Panchromatic Nega- 
tive Safety Film show- 
ing effect of tempera- 
ture during exposure. 



during exposure on the reciprocity- 
failure characteristics. The group of 
reciprocity-failure curves in Fig. 8 
illustrate the kind of change that takes 
place when the temperature is varied 
over a wide range. It is seen that as 
the temperature is lowered there is a 
continual increase in the reciprocity 
failure at high intensities and a decrease 
at low intensities, which reaches a 
maximum at about 40 F. At the 
intensities encountered in motion picture 
photography, it is apparent that the 
sensitivity of the film is directly related 
to its temperature at the time of exposure. 
Although the range in temperatures 
shown in this figure is great, it is quite 
possible that in location work and back- 
ground photography the temperatures 
encountered may approach these ex- 
tremes perhaps from 40 F to 120 F. 
In Fig. 9 is shown a family of D-log E 
curves for Eastman Plus-X Panchromatic 
Negative Safety Film exposed at 1/100 
sec at the temperatures indicated ( 50 
F to 200 F). It will be noted that in 
addition to a change in sensitivity, which 
in this case is greater than a factor of 2, 
there is also a drop in gamma as tempera- 
ture is reduced. The magnitude of 
these changes is fairly typical of negative 
materials, although with some films, 
notably infrared-sensitive films, the 
change in sensitivity is appreciably 



greater. Over the range of tempera- 
tures normally encountered in studio 
photography, these changes are, of 
course, negligible. 

It is regrettable that it is not possible 
at this time to present a more compre- 
hensive treatment of reciprocity-law 
failure at the extremely high intensities 
which are occasionally encountered 
in ultra-high-speed photography. This 
lack of data is a consequence of the very 
great experimental difficulty of making 
complete and reliable measurements at 
exposure times shorter than about 
10~ 6 sec. It is expected that suitable 
apparatus will be available in the near 
future with which we can extend our 
knowledge into this region. 

Discussion 

John H. Waddell (Wollensak Optical Co.): Mr. 
Tapper, the things that you illustrated in ref- 
erence to temperature were extremely interest- 
ing. However, I don't believe that the tempera- 
ture range was quite enough. We would like 
to see from 80 to about +180, the reason 
there being that at White Sands Proving Ground 
if you take a camera which is painted black and 
allow it to stand in the sun from four to six hours, 
that camera cannot be touched by hand. In the 
meantime, the film has been cooking all of that 
time, but it will produce color pictures on Koda- 
chrome at speeds in daylight up to 1000/sec with 
full exposure, so that the comments you have 
made on temperature range are extremely 



28 



January 1953 Journal of the SMPTE Vol. 60 



valuable. But we would like to see it extended 
just a little bit more. 

Mr. Tupper: There is some information avail- 
able for temperatures both higher and lower 
than those indicated in the graph which showed 
the effect of temperature on reciprocity-law 
failure. As the temperature of the film is lowered 
to very low levels (300 F) the reciprocity 
curve flattens out and the sensitivity of the film is 
very low at all intensities. When the tempera- 
ture is raised to 200 F the change in reciprocity- 
law failure follows the same general pattern as 
shown in the graph, that is, an increase in effec- 
tive sensitivity at high intensities and a decrease 
at low intensities. In the case of the color films, 
the color balance is affected only if the changes in 
reciprocity-law failure are different for each of 
the emulsion layers. I believe that the changes 
follow the same pattern in each of the component 
emulsions. 

Mr. Waddell: I am not particularly worried 
about color balance at those speeds, because 
I'm interested in comparative color rather than 
true color. However, there is one other question 
that I would like to ask: what is the effect on 
film speed with variation in temperature? Can 
one begin to assume that the old chemical rule 
of doubling of film speed for every 10 degrees rise 
in centigrade holds generally true or even as an 
approximation? 

Mr. Tupper: I pointed out that the magnitude 
of the change in sensitivity with temperature 
depends upon the intensity level at which ex- 
posures are made. In the region of optimum 
intensity the changes in sensitivity with tempera- 
ture are not very great, but at very high and 
very low intensities the changes are considerable. 
However, even at the extremes I expect that a 
factor-of-two change in sensitivity represents a 
greater temperature difference than 10 G with 
normal photographic materials. It is more likely 
that a change of 20 C or greater is required to 
produce a factor-of-two change in sensitivity. 

Kenneth Shaftan (J. A. Maurer, Inc.) : Some time 
ago Dr. O'Brien's group at Rochester were 
working on the matter of reciprocity failure and 
they found some specific information related to 
dye sensitization. I wonder if you have carried 
out any further work in that direction? 

Mr. Tupper: There has been some work done 
by J. H. Webb on this problem, but I believe 
he has as yet nothing definite to report. 

Mr. Shaftan: The second question relates to 



the means or measuring exposure durations 
shorter than 10~ 5 sec. At 10~* sec there has 
been a considerable amount of work with regard 
to sound on film, but what is happening in the 
realm beyond that? 

Mr. Tupper: The only apparatus which is 
now available for our measurements at very 
short exposure times was designed and built by 
J. H. Webb. This apparatus consists of a high- 
speed rotating drum which carries the film past 
the projected image of an illuminated narrow 
slit. We can record to about 10~* sec with 
fast negative films. The curves which I have 
just shown were based on exposures made on 
this apparatus. We are at present considering 
the design of a sensitometer which will provide 
exposures at 10~ 7 or possibly 10~ 8 sec. 

Brian O'Brien (University of Rochester): Perhaps 
I can elaborate a little on this reciprocity failure. 
I think Mr. Tupper knows about the work we 
have done. For those of you who are interested 
in practical high-speed photography, as long 
as you keep to relatively slow speeds below a 
million per second you don't have to worry 
at any wavelength or any combination to be 
found. So let me set your minds at ease at once, 
unless you're going to go to really very high 
speeds. 

The unusual effects we find are limited to 
exposures less than Iju/sec and, therefore, to 
cameras running faster than 1,000,000 frames/ 
sec. Our experiments thus far are limited to 
exposure times down to 1 X 10~ 7 sec. At about 
10~ 6 sec a new reciprocity failure, not taken into 
account by the usual Gurney and Mott theory 
or more recent solid-state modifications of it, 
occurs for those regions of the spectrum to which 
a photographic emulsion is dye-sensitized. It 
does not appear to occur at all for regions in the 
blue where the natural sensitivity of the silver 
halide grains is found. Thanks to fine coopera- 
tion from the Kodak Research Laboratory which 
has prepared special emulsions for our use, it 
has been possible to test this phenomenon to our 
present experimental limit of 10~ 7 sec of exposure. 
It is a very interesting problem from the theo- 
retical standpoint, and we hope soon, with a new 
camera, to reach exposures as short as 10~ 9 sec 
to further explore this type of reciprocity failure. 
Let me emphasize, however, that these are very 
short exposures, and I am speaking of a time 
range which is quite beyond any ordinary high- 
speed photography. 



L. Tupper: Reciprocity-Law Failure 



A Method of Lighting Large Fields 

for High-Speed Motion Picture Photography 



By HARRY R. GLASON 



JL\T THE National Advisory Committee 
for Aeronautics we were confronted, 
some time ago, with the problem of 
lighting and photographing at 6000 
frames/sec, the path of a missile 6 in. 
in diameter, traveling for a distance of 
30ft. 

Calculations indicated it would re- 
quire 400 No. 2 photofloods, drawing 
1800 amps of current to furnish the 
6,000,000 1m required. As this was 
impractical, we turned to the use of the 
photoflash bulb. 

One No. 11 and one No. 50 fired at 
the same instant maintained the neces- 
sary light, from 15 to 50 msec after the 
missile's flight. A microswitch installed 
in the gun's breach, set off the flashbulbs 
as soon as the missile started moving. 
The delay in the No. 11 flashbulb was 
calculated to coincide with the time 
required for the missile to emerge from 
the muzzle. 



Presented on October 10, 1952, at the 
Society's Convention at Washington, D.C., 
by Harry R. Clason, National Advisory 
Committee for Aeronautics, Langley Field, 
Va. 



We next built a delay circuit to fire 
flashbulbs with fixed amounts of delay 
between bulbs. Existing data on flash- 
bulb characteristics indicate that eight 
No. 31 focal-plane bulbs fired in rapid 
succession with 55-msec delay between 
each bulb will give a light variation 
between 1.15 and 1.45 million lumens, 
or only 26% variation, for a total dura- 
tion of 410 msec. This is long enough 
to expose 60 ft of film at 6000 frames/sec. 

The exposure guide number at 6000 
frames, with a 64 speed rating film is 16. 
This means we can take 6000 frames/ 
sec, with lights 4 ft away, at an //stop 
of 4. 

Eight No. 50 flashbulbs fired with 
20-msec delay, will give a light variation 
between 4 and 6.5 million lumens, or 
60% variation for a total duration of 
150 msec. This is long enough to 
expose 25 ft of film at 7000 frames/sec. 

The exposure guide number for the 
No. 50 at 7000 frames, with a 64 speed 
rating film is 32. This means we can 
take 7000 frames/sec, with the lights 
8 ft away, at an //stop of 4. 



30 



January 1953 Journal of the SMPTE Vol. 60 



X-Ray Motion Picture Camera 
and Printer for 70mm Film 

By S. A. WEINBERG, J. S. WATSON, and G. H. RAMSEY 



A cinefluorographic motion picture camera and reduction printer using 70mm 
perforated negative film are described. The camera drive mechanism 
permits camera speeds up to 15 frames/sec. Prints can be made either on 
35mm or 16mm positive film. 



JL HE APPARATUS here described is 
intended primarily for making x-ray 
motion pictures 1 on 70mm negative. In 
this field 70mm film has some obvious 
advantages over 35mm film. It provides 
better reproduction of subject detail and 
is also easier to study frame by frame. 2 

The new 70mm camera (illustrated in 
Fig. 1 and Fig. 2) was built around an 
already existing negative film, Eastman 
Linagraph Ortho 70mm. This film can 
be obtained perforated to American 
Standard dimensions, and although the 
perforations are not specifically intended 
for use in the motion picture field, we 
have had no reason so far to regret our 
choice of the ready-made film. The 
center-to-center distance between per- 

Presented on October 9, 1952, at the 
Society's Convention at Washington, D.C., 
by Sydney Weinberg, Dept. of Radiology, 
University of Rochester School of Medicine 
and Dentistry, 260 Grittenden Blvd., 
Rochester 20, N.Y. This project was 
supported in part by a research grant from 
the National Heart Institute of the Na- 
tional Institutes of Health, Public Health 
Service. 



forations is 0.234 in. The vertical 
dimension selected for our camera frame 
is 9 perforations or 2.106 in., less 0.025 
in. for the frame line. Frame width is 
2.25 in., providing a frame which is 
more nearly square than the standard 
motion picture frame and therefore 
more in line with routine x-ray viewing 
practice. (See Fig. 3.) 

Because of the long travel (more than 
2 in.) of the film pulldown mechanism, 
and also because of the large mass of 
the 70mm film as compared with per- 
foration size, it seemed advisable to 
plan for multiple pulldown teeth on 
each side. The Bell & Howell 35mm 
speed movement has the special merit 
of permitting multiple teeth without 
adding unduly to the weight of the 
vertically moving parts, and for this 
reason it was chosen as the model for our 
70mm film movement. We found in the 
end that 6 teeth on each side were 
necessary. A pilot model of the camera 
with 5 teeth produced small fracture 
marks in the perforations, but in the 
finished camera the addition of an extra 
tooth on each side provided the remedy. 



January 1953 Journal of the SMPTE Vol. 60 



31 




Fig. 1. 70mm camera equipped with 150-mm //0.85 Leitz objective 
and 400-ft lead-covered film magazine. 




32 



Fig. 2. Inside view of 70mm camera with film gate open. 
January 1953 Journal of the SMPTE Vol. 60 








Weinberg, Watson and Ramsey: X-ray Camera 



33 



The teeth of the pulldown are spring- 
loaded and extend into the perforations 
just enough to give a full bearing surface. 
During the up-stroke, while the film is 
stationary, the rounded top edges of the 
teeth slide over the perforations much 
in the fashion of a stick over a picket 
fence. At the start of the down-stroke 
the squared lower edges of the teeth 
engage the perforations. The teeth 
never extend or retract more than 0.007 
in. 

Pulldown and dwell periods are equal. 
This ratio is convenient for synchronizing 
the camera with the 60-cycle pulsed 
x-ray output. The camera drive is a 
J-hp synchronous motor linked to the 
camera through a gearbox maintaining 
synchronism through a choice of film 
speeds of 15, 1\ and 3j frames/sec. 
Incorporated in the camera drive is a 
commutator which triggers an electronic 
contactor, which in turn interrupts the 
power supply to the x-ray generator 
during film transport. This not only 
spares the patient unnecessary exposure 
to x-radiation but also extends the time 
during which the x-ray tube may be 
energized for any one examination or 
series of examinations. 

In order to minimize vibration the 
non-moving parts of the camera are of 
sturdy construction. The "box" without 
lens, magazine, or motor drive weighs 
60 Ib. Fully assembled on a rigid lathe 
bed, and with the fluorescent screen 
assembly in place, the unit weighs about 
250 Ib. 

It will be remembered that the 
Bell & Howell speed movement has a 
flat aperture plate without any curve 
above or below the aperture, and that 
the "pressure" plate is also flat and is 
not pushed by springs or cam action 
against the back of the film to flatten 
it, as in other film movements. Tests 
indicated that 70mm film in a movement 
of this sort does not in fact lie flat, but 
bulges centrally into the aperture, some- 
times by as much as 0.05 n. This 



deviation from a flat plane is not constant 
but varies from frame to frame. 

A lens of small aperture might have 
sufficient depth of focus to take care of 
the bulging, but considering that de- 
sirable Senses for cinefluorography have 
apertures of //0.85 and more, it is 
apparent that only a few ten-thousandths 
of an inch variation will result in out-of- 
focus areas on the film. 

The readily available solution to this 
problem was to incorporate a compressed 
air chamber in the camera so that a 
cushion of air would be formed between 
the rear element of the lens and the film 
emulsion, thus flattening the film against 
the pressure plate. Our intermittent is 
designed so that the clearance between 
the aperture plate and the pressure plate 
is adjustable. This permits varying the 
air-escape passage to get a satisfactorily 
uniform flow of air and/or uniform air 
pressure. Too narrow an escape chan- 
nel produces a buildup of air pressure 
which immobilizes the film against the 
pressure plate, resulting in torn per- 
forations, while too wide a channel 
permits a fall of pressure and bulging of 
the film. 

From our experience it is evident that 
a number of air pressure-air escape 
combinations will permit satisfactory 
operation. The present separation of 
aperture from pressure plate is 0.010 
in., or 0.004 in. greater than the thick- 
ness of the film. The air stream is 
continuous and is finally vented through 
a light-tight trap in the body of the 
camera. Air pressure is about 2 psi. 
An oil, dust and moisture filter is 
incorporated in the air line, and a 
solenoid valve automatically opens the 
line when the camera is started. A 
diagram of the air chamber is shown in 
Fig. 4. 

The fact that the film does not come 
in contact with the aperture plate 
eliminates scratching of the emulsion, 
and although fine scratches sometimes 
appear on the back of the film (despite 
the high polish of the pressure plate) 



34 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 4. Schematic of camera showing air-pressure chamber and 
air-escape passage (exaggerated). 



these are so small in relation to the 
70mm frame as to be imperceptible on 
contact or reduction prints made with 
diffused light. Registration has been 
good enough so that we have not felt 
inclined to measure any variations that 
might be present. 

The 70mm Optical Printer 

The 70mm printer head is a scaled-up 
version of a conventional camera move- 
ment with spring-loaded pressure plate 
and side rail, and without separate 
register pins. The pulldown mechanism 
has two teeth on each side. There is 
no doubt that one tooth would be 
sufficient, but the extra tooth is included 
as an added assurance of good registra- 
tion. At the completion of the pulldown 
stroke the teeth are not withdrawn from 
the perforations until exposure is com- 
pleted. Tolerances have been kept 
extremely close. We theorized that four 
teeth instead of two would tend to 
average out possible inequalities between 
perforations. 

Prints for projection are generally 
made on 16mm fine-grain positive, but 
sometimes on reversal duplicating film 



when negative-image prints are desired. 
Master positives are made on 35mm 
duplicating positive. Changing from 
one film size to the other merely involves 
a change of cameras and of film-to-film 
and lens-to-film distances. Provision 
is made for printing at various magni- 
fications and for vertical and horizontal 
shifts in image position. 

The printer turns over at a slow rate 
(2 frames in 3 sec). This is not in- 
conveniently slow considering the rela- 
tively short length of individual cine- 
fluorographic scenes. The projector 
head and camera (shown in Fig. 5) can 
be operated separately, in stop motion, 
through indexing type clutches. Pro- 
jector or camera can be run forward or 
backward. They can also be run 
continuously with automatic printing of 
two positive frames for each frame of 
negative. In this way action originally 
photographed at n frames/sec can be 
slowed down in the print to 2n frames/ 
sec. 

Other 70mm Cameras, Past and Present 

Probably the earliest 70mm motion 
picture camera was the Mutograph of 



Weinberg, Watson and Ramsey: X-ray Camera 




36 



January 1953 Journal of the SMPTE VoL 60 



Herman Casler (1895), an example of 
which is on view at George Eastman 
House. The camera was motor-driven 
at a speed of 40 frames/sec, had an 
intermittent of the "broken-roller" type, 
and used unperforated 70mm roll film, 
punching two register holes in each frame 
as it came to rest. The Mutograph 
negatives were printed on paper for 
showing in Mutoscope peep-show ma- 
chines. Another 70mm camera, this 
time for perforated film, was built by 
the Lumieres. Contact prints from the 
70mm negatives were projected on a 
giant screen (16 X 21 m) at the Paris 
Exposition of 1900. Nearly 30 years 
later, during the wide-film craze of 1929, 
a pair of 70mm cameras was constructed 
in connection with the Fox "Grandeur 
Film" project. The Fox camera aper- 
ture was of panoramic shape (22.5 X 48 
mm) with provision for a 10-mm sound 
track. 3 

The only contemporary camera com- 
parable with our own is believed to be 
the 70mm "high-speed sequence camera" 
manufactured by the Charles A. Hulcher 
Company of Hampton, Va. 4 At present 
the Hulcher camera is not adapted to 
the making of x-ray motion pictures 
because of its uncertain registration of 
frames and because of the relative 
shortness of its dwell period. 



References 

1. S. A. Weinberg, J. S. Watson, and 
G. H. Ramsey, "X-ray motion picture 
techniques employed in medical diag- 
nosis and research," Jour. SMPTE, 59: 
300-308, Oct. 1952. 

2. J. S. Watson, S. A. Weinberg, and G. 
H. Ramsey, "A 70mm cinefluorographic 
camera and its relation to detail," 
Radiology, Dec. 1952. 

3. Carl Louis Gregory, "The early history 
of wide films," Jour. SMPE, 14: 27, 
Jan. 1930. 

4. Charles A. Hulcher, "70mm high-speed 
sequence camera," presented at the 
Society's Convention on October 10, 
1952, and planned for early publication 
in the Journal. 

Discussion 

William H. Unger (Elliot, Unger, & Elliot, 
Inc.): I'd like to know if you relieve the 
air pressure during the pulldown cycle. 

Mr. Weinberg: The air pressure is con- 
stant. It is not pulsed. We have a small 
escape port in the side of the camera. 
Does that answer your question? 

Mr. Unger: Yes. Were the films 
double-printed on alternate frames to 
bring them to approximately normal 
speed or is all the action speeded up? 
In other words, did you print your nega- 
tive one frame to one? 

Mr. Weinberg: There's a large variation 
in all these films. I couldn't possibly 
point them out now. Some of them were 
2:1, some 3 : 1 and some straight prints. 



Weinberg, Watson and Ramsey: X-ray Camera 



Application of Wide-Angle Optics to 
Moderately High-Speed Motion Picture Cameras 



By H. E. BAUER and A. W. BLAKE 



The Douglas Aircraft Co. has gained considerable experience in the applica- 
tion of wide-angle optics to certain motion picture cameras, particularly at 
200 to 500 frames/sec. The field of view used ranged from 140 to 160 full-cone 
angle, and aperture settings of the order of f/1.5 were obtained. Further, 
the extreme depth of field of the optical system was found to be a very useful 
feature. Experiences concerning the development of these wide-angle 
cameras are discussed. 



I 



N 1948, the Douglas Aircraft Co. was 
presented with a very intriguing problem 
concerning the development of an in- 
strumentation system that could record 
and thus provide the means of recon- 
structing the relative approach histories 
of certain high-speed objects arriving in a 
somewhat random fashion about an 
aerial target. A variety of systems be- 
sides photography, based on acoustic, 
radar and other physical principles, were 
considered; however, for several rea- 
sons, many beyond the scope of this pa- 
per, it was decided to pursue the de- 
velopment of a photographic system in- 
tended to yield film records suitable for 
eventual photogrammetric analysis. This 
proved to be a wise decision and the en- 
suing program was eminently successful. 



Presented on October 9, 1952, at the 
Society's Convention at Washington, D.C., 
by H. E. Bauer who, with his coauthor, 
A. W. Blake, is of the Engineering Dept., 
Douglas Aircraft Co., Inc., Santa Monica, 
Calif. 



Due to reasons of security classifica- 
tion, a further general discussion of the 
complete photographic instrumentation 
and its use must be of a limited nature in 
this paper; however, some of the un- 
classified components of this system can 
be revealed. Wide-angle optics as ap- 
plied to moderately high-speed motion 
picture cameras will be discussed. 

Objectives 

An analysis of the overall system re- 
quirements brought forth the following 
general specifications as design objec- 
tives for wide-angle motion picture cam- 
eras. 

(1) Field of view to be at least 142 
full-cone angle; 

(2) Relative aperture to be near //1. 5, 
allowing an exposure time of the order 
of one millisecond or less for color film; 

(3) Sampling rate to be of the order of 
200 to 500 frames/sec; 

(4) Visual range for a significant ob- 
ject to be of the order of 250 ft; 



38 



January 1953 Journal of the SMPTE Vol. 60 






(5) Reliable operation and reasonable 
service life to be achieved ; 

(6) A means of recording a time base 
with a time resolution of the order of one 
millisecond to be provided. 

In addition to the above, it was de- 
sired that the evaluation of radial meas- 
urements of significant images from the 
film records be accomplished with an 
accuracy of f. 

Previous and Concurrent Developments 

The initial phase of the program for 
the development of optical components 
for the instrumentation system began 
with a survey of known wide-angle lens 
assemblies. The results of the survey 
were not encouraging, for in general the 
available lens designs proved to be inade- 
quate in several respects. Among the 
group of "corrected" wide-angle lenses 
investigated, field angles were usually 
less than 90 and relative apertures not 
much better than //6.8. Typical lenses 
in this category include : 

(1) Eastman Ektar, 75, //6.3; 

(2) Bausch & Lomb Metrogon, 85 J, 
//6.8; 

(3) Wollensak Raptar, 84, //6.8 
Although these are all very fine lenses 

and there are many others, they did not 
meet the primary objective of a view 
field of 142 and a large enough aperture 
to allow millisecond exposure. 

The results of investigations into the 
more extreme wide-angle lenses showed 
that most of them were inapplicable with 
respect to available relative aperture. 
The most extreme wide-angle lens that 
is corrected for rectilinear rendering is 
the Goerz Hypergon with a focal length 
of 60 mm. This lens covers a field of 
135 at f/22 on a 5 X 7 in. plate. 
Another type of lens with similar field 
coverage is the//9 Goerz Dagor made by 
Carl Zeiss. Greater coverage is offered 
with the Robin Hill lens made by Beck 
of England; it yields 180 at //1 6. 
Another very interesting adaptation of 
the Robin Hill design is made by Zeiss 
with a fair degree of correction. It has 



a view field of 210 at //6.8, a focal 
length of 1.6 cm, and covers a 2\ in. 
square plate. 

Optics 

Some very interesting and useful re- 
sults may be obtained by application of 
the basic philosophy of the Robin Hill 
lens design the placing of a supple- 
mentary lens of negative characteristics 
in front of a positive objective. This is 
often done when the back focus of a 
standard wide-angle lens is too short to 
accommodate extra apparatus in front 
of the film plane, such as the shutter of a 
cine camera, the beam-splitting pri:m in 
a Technicolor camera, or the rotating 
prism in a continuous film-flow, high- 
speed cine camera. The principles in- 
volved are shown in a most elementary 
schematic fashion in Fig. 1. This 
negative-positive lens arrangement, sim- 
ilar to a reversed telephoto lens system, 
stimulated considerable interest since 
the survey of other known lenses had 
shown little promise. 

A brief theoretical appraisal of the de- 
sirability of using an uncorrected nega- 
tive lens was made. It revealed that 
this type of lens system would offer no 
benefits to photography other than wide- 
angle coverage and the use of a high- 
speed positive lens. The application of 
standard techniques for complete optical 
correction to such an arrangement 
would most likely lead to a long and 
costly development program with no 
assurance of successful results. How- 
ever, it appeared that a few partial cor- 
rections could be applied which seemed 
likely to bring this system within the 
threshold of acceptable performance for 
the requirements of the unique instru- 
mentation system being developed. I 
the negative-positive lens couplet could 
be readily adapted to the concurrent 
high-speed motion picture camera de- 
velopment program which had been in- 
itiated along with the optical study, then 
further detailed consideration would be 
scheduled. 



Bauer and Blake: Wide-Angle Optics for High-Speed 



39 






NEGATIVE 
/ FIELD LENS 



POSITIVE 
OBJECTIVE LENS 




IMAGE OF PICTURE 



VIRTUAL IMAGE 
OF NEGATIVE LENS 



FILM PLANE 



Fig. 1. Wide-angle optical system schematic. 




OJ47 



CIRCULAR MIRROR 



PACIFIC OPTICAL 
- 1 IN. COATED 
NEGATIVE DOUBLET 
( PLASTIC ) 



WOLLENSAK I IN., f /I.5 
COATED RAPTAR LENS 



Fig. 2. Wide-angle optical system design layout. 



There was in use at Douglas, a two- 
element wide-angle negative reducing 
lens which was employed for periscopic 
visual inspection of inaccessible fields of 
view. Experiments were made based on 
the application of this type of nega- 
tive scanning lens used in couple with 
standard camera objective lens and the 
results were rather impressive. A layout 
of this optical system is shown in Fig. 2, 
and Fig. 3 shows a black-and-white re- 



production of typical color pictures 
taken with these optics using 35mm film. 
Since this system was filled with all sorts 
of aberrations, the pictures did not meet 
the high standards of first-rate photo- 
graphic rendering; however, for our 
unique requirement, the results were 
more than adequate. The required 
field angle was reached; in fact, it was 
exceeded, as field angles of 160 were 
readily attained, and the relative aper- 



40 



January 1953 Journal of the SMPTE Vol. 60 




TIMING LIGHT 
IMAGES 



.750" DIA. 
PICTURE 
REPRESENTING 
152 FIELD 
CONE ANGLE 



Fig. 3. Typical pictures taken with wide-angle optics, using color film in 35mm 
camera; this is an aerial view of a desert village from an altitude of 2500 ft true, 
with camera rolled 22 and pitched down 13. 



ture was found to be slightly less than 
that available from the camera objective 
lens. Thus, a relative aperture close to 
//1. 5 was realized. 

The question may here arise as to how 
it is possible to obtain a picture that has 
any degree of sharpness from an optical 
design which has no apparent correction. 
It should first be recalled that the optical 
design is intended for a very singular 
application. When faced with the need 
of a high-speed extremely wide-angle lens 
system, some desiderata had to be given 
up. Even a brief perusal of the design 



under discussion will reveal defects of 
astigmatism and curvature of field as well 
as distortion. Distortion can be cali- 
brated and a requirement of true recti- 
linear propagation can thus be circum- 
vented. To account for astigmatism, 
and curvature of field, a convergent lens 
would have to be introduced into the 
field-lens elements, but such an addition 
would cut down the extremely wide 
angle and the speed. Actually, curva- 
ture of field is minimized by the extreme 
depth of field of this very short 0.147-in. 
focal length system; objects practically 



Bauer and Blake: Wide-Angle Optics for High-Speed 



41 




42 



January 1953 Journal of the SMPTE Vol. 60 



in contact with the front lens element are 
in sharp focus, as are objects at extreme 
range. Although not corrected for curv- 
ature of field, the effects are considerably 
reduced by this great depth of field. 
The caustic image surface produced by 
astigmatism remains as the major detri- 
mental factor in the system, and this 
effect is only partially minimized by the 
depth of field. In order to maintain the 
aperture rating //I. 5, with the 142 
field coverage, allowances will have to be 
made for the unavoidable residual aber- 
rations encountered in extremely wide- 
angle optics. 

Wide-Angle Optical Assembly 

The optical assembly which converts a 
motion picture camera into a wide-angle 
(142 full-cone angle) camera, with pro- 
vision for time correlation and suitable 
for photogrammetric application, has 
the following components: 

(1) A conventional cine objective lens 
of high relative aperture ; 

(2) An external means of adjusting the 
objective lens diaphragm; 

(3) A supplementary negative field 
lens; 

(4) A lamp cartridge repeater clock 
containing a battery of 10 timing lights; 

(5) An annular mirror to reflect the 
timing lights in proper focus onto the 
marginal section of the film frame ; 

(6) Provisions for protecting the optical 
system against fogging or icing; 

(7) A fiducial mark on the field lens 
within the focus range of the objective 
lens. 

Design considerations of such a wide- 
angle optical assembly, in which the 
camera objective lens is focused on the 
virtual image of a negative field lens, are 
based on three fundamental variables: 
effective focal length of the negative lens, 
effective focal length of the positive lens, 
and the separation of the two lenses. 
The combined focal length of the assem- 
bly is a function of the above variables 
and many combinations of these were 



studied and tested. Considerations oi 
field coverage, magnification, component 
packaging and quality control led to a 
production design shown in cutaway 
form in Fig. 4. For use with 16mm 
motion picture cameras, a 25-mm //1. 5 
Wollensak Raptar objective was chosen, 
while for 35mm cameras, a 50-mm//1.5 
Wollensak Raptar Dumont objective 
was used. 

The negative doublet is a 25-mm 
focal length plastic lens; it will be dis- 
cussed in detail later. The lens separa- 
tion is 6.80 in. for 16mm cameras and 
7.37 in. for 35mm cameras. Thus, for 
16mm film, an 0.30-in. diameter circular 
image (representing a usable 142 field 
cone angle) is registered while a 0.75-in. 
diameter image (representing 152) 
results when a 35mm film is used. In 
either case, sequential images on the film 
are tangent to each other to utilize 
maximum magnification. 

Although most experimental versions 
of this wide-angle optical assembly were 
designed and developed by Douglas 
engineers, the production articles as 
shown in Figs. 4 and 5 are the result of a 
joint design effort by Douglas and 
Wollensak engineers. The production 
lens assemblies are now manufactured by 
the Wollensak Optical Co. in compliance 
with Douglas specification drawings. 

The plastic lens is manufactured by 
the Pacific Optical Co. of California and 
the degree to which quality control has 
been attained is an achievement in itself. 

Plastic Wide-Angle Reducing Field Lens 

The machined and polished negative 
field lens used with considerable success 
in this wide-angle package is made from 
Plexiglas which has an index of refrac- 
tion of 1 .49 for the sodium D lines. The 
front 4-in. diameter element is convex- 
concave and the rear 3-in. diameter ele- 
ment is bi -concave. The lens design in- 
corporates no built-in aperture stop but 
there is a measurable entrance pupil of 
1 .06 in. in diameter and an exit pupil of 
1.82 in. in diameter. The mean focal 



Bauer and Blake: Wide-Angle Optics for High-Speed 



43 




Fig. 5. Production 
wide-angle optical as- 
semblies applied to 
high-speed cine cameras: 
top, 35mm Bell & Howell 
Ultra-Speed; middle, 16- 
mm Wollensak Fastax; 
bottom, 16mm Bell & 
Howell Filmo. 








44 



January 1953 Journal of the SMPTE Vol. 60 




length is a minus one inch for this com- 
bination. 

Even though there are no designed 
corrections made for the aberrations in 
this field lens, there are some compensa- 
tions (a number of which have already 
been discussed) that can be designed into 
the complete lens assembly to minimize 
the effect of these aberrations. 

To increase light transmission by re- 
ducing reflections, the plastic lens was 
given a standard coating with a fluoride 
compound. This coating also dimin- 
ished the surface light-scattering effect 
(haze) prevalent in plastics, which along 
with reflections, tends to cut down pic- 
ture contrast. The baking process used 
to harden the coating on glass lenses 
cannot, of course, be used with plastics, 
yet the unbaked coating has been found 
to be quite durable. 

Glass Wide-Angle Reducing Field Lens 

In an effort to explore the possibilities 
of improving resistivity against abrasion 
and, possibly, also the optical properties, 
a version of the negative doublet was 
designed and made of glass. The ex- 
terior dimensions were essentially the 
same as those of the plastic lens. 

Studies of the inherent qualities of the 
glass doublet in comparison with its 
plastic counterpart were made. It ap- 
pears that some advantage may be 
gained from the use of the glass lens with 
respect to light-scattering and crazing, 
but any benefits in resolving power 
failed to show up in tests. Pictures 
taken with a glass wide-angle optical 
assembly compared to those taken with a 
plastic assembly show little, if any, im- 
provement. The small advantages of 
the glass lens are outweighed by the 
relative cost of the article when compared 
with the end product. 

Time-Correlation and Repeater Clocks 

One of the objectives of the program 
was to develop a means of referring the 
camera's film record to a common time 
base with a time resolution of the order of 



one millisecond. This was desirable for 
several reasons, most of which are beyond 
the scope of this presentation. How- 
ever, one important reason was to pro- 
vide a means of correlating time between 
two or more cameras recording the same 
event, thus avoiding the need for a costly 
exposure synchronization system. 

This objective was met by the de- 
velopment of an Electronic Master Time 
Clock (syncopic binary time code gener- 
ator) which provides accurately timed 
electrical pulses eventually to be pre- 
sented to the view field of all related 
cameras in the form of an array of 
flashing neon lamps called Repeater 
Clocks. The resultant images on the 
film records form a unique syncopic bi- 
nary code, so constructed that only one 
digit changes at any step, thus avoiding 
misinterpretation when carry-over, or 
"double" exposure, occurs during a 
camera exposure cycle. 

Cameras 

Although this paper deals primarily 
with the development of a wide-angle 
optical system of large relative aperture, 
it may be of interest to mention another 
development program that was initiated 
concurrently concerning the selection of 
motion picture cameras in the speed 
range of 200 to 500 frames/sec intended 
to use these wide-angle optics when de- 
veloped. 

In formulating the design philosophy 
guiding the development of suitable 
types of motion picture cameras into 
wide-angle moderately high-speed cam- 
eras, suitable visual range, sampling 
rate, reliable field operation and reason- 
able service life were considered. Stud- 
ies were made comparing the relative 
merits of moderately high-speed 16mm 
cameras, as well as 35mm models. The 
results of the studies led to the selection 
of the Wollensak Fastax and the Bell & 
Howell Filmo cameras as two types suit- 
able for development. In the 35mm 
field, only the Bell & Howell Ultra- 
Speed was considered suitable. These 



Bauer and Blake: Wide-Angle Optics for High-Speed 



45 



cameras, shown in Fig. 5, illustrate the 
production lens assembly using the plas- 
tic field lens. 

The Fastax camera was modified for 
best operation in the 400- to 500-frames ' 
sec range. It may be of some interest 
here to note that it was a slight task to 
slow the camera down for reliable opera- 
tion at these relatively low speeds. 

In the field of 16mm intermittent 
motion picture cameras, the survey of 
available types led to laboratory tests 
with the GSAP (Gun Sight Aiming 
Point) and the Bell & Howell Filmo 
Model 70. The well-known family of 
GSAP cameras was found to be easily 
modified to accept wide-angle optics of 
160 full-cone angle, and satisfactory 
performance was experienced when the 
cameras were run at 64 frames sec. 
which is within their design limitations. 
However, attempts to "hop up" the 
GSAP to yield relfable frame rates up to 
200 frames/ sec were soon terminated. 
Although limited operation at these ex- 
cessive frame rates was occasionally 
attained, the camera then became so 
unreliable that success with future de- 
velopments seemed very unlikely. Lab- 
oratory tests with the Filmo yielded re- 
sults with frame rates up to 230 frames/ 
sec and best reliability in the 1 80- to 200- 
frames/sec range. 

It was through the cooperative efforts 
of Training Aids Inc. (special California 
representative of the Bell & Howell Co. 
of Chicago), the Bell & Howell Co., and 
Douglas engineers that the Model 70 
Filmo was modified for reliable operation 
at 200 frames/sec and adapted to accept 
the same standard wide-angle optical 
assembly used on the Fastax Camera. 
Its photographic performance surpassed 
expectations and is matched only by the 
Ultra-Speed camera in visual range and 
photographic sharpness. 

In the realm of 35mm intermittent 
motion picture cameras, the Bell & 
Howell Ultra-Speed was an obvious 
choice, for it had already been developed 
as a 200-frames/sec camera. Later in 



the program, it was modified to meet the 
system's requirements. The modifica- 
tion task here was simply that of attach- 
ing a wide-angle optical package. 

Film 

It became apparent in the early phases 
of the development program that pic- 
tures taken with color film were far more 
satisfactory than with black-and-white. 
The optical system under discussion pre- 
sents some difficulties with chromatic 
differences of magnification. This situ- 
ation becomes more critical at the edge 
of the field of view ; however, a more dis- 
tinct image is apparent on color film such 
as Kodachrome because there are three 
different color-sensitive layers to register 
monochromatic separate images. The 
resultant fringing actually produces a 
colorful outline that is considerably 
easier to distinguish than the fuzzy 
grayed outlines produced on black-and- 
white film. 

In an attempt to improve the quality 
of film records, experiments were made 
with filters so as to produce, in a sense, 
more monochromatic light. Test re- 
sults using filters with black-and-white 
film were not very promising; however, 
results from using filters with Koda- 
chrome film were more gratifying. 

Daylight, Type "A," and Commercial 
Kodachrome films were tried with suit- 
able filters. Daylight and Type "A" 
being inherently more contrasty than 
the commercial emulsion were found to 
give a higher overall resolving power, 
since resolving power increases with 
contrast. As previously discussed, the 
more monochromatic the registering 
light, the better the film records. With 
this in mind, tests were made to compare 
Type "A" used with its compensating 
filter, and Kodachrome Daylight film 
used with a normal haze filter. It 
appears from tests that the compensating 
filter for the Type "A" film, being heavier 
(absorbing more light at the blue portion 
of the spectrum) than the haze filter for 



46 



January 1953 Journal of the SMPTE Vol. 60 



Daylight film, does yield some slight 
advantage. 

Resolving Power 

As expected, this optical system suffers 
from astigmatism. This effect is noticed 
in resolving power tests in an interesting 
manner. In the paraxial region, maxi- 
mum resolving power of Kodachrome is 
almost obtained. Radial resolving 
power drops off very little from the center 
to the picture edge, while tangential re- 
solving power drops off at a much 
greater rate. 

Conclusions 

The development program, guided by 
the broad design objectives previously- 
discussed, was successful in producing 
several versions of moderately high-speed 
motion picture cameras equipped with 
wide-angle optics. 

The wide-angle optical system as de- 
veloped during this project yields excel- 
lent results for the purpose intended. 
These results were achieved by a careful 
program of minimization of the system's 
optical deficiencies and exploitation of 
its capabilities. 

In practice, the photograrnmetric 
analysis of camera film records was 
readily accomplished to the degree of 
accuracy desired, and the photographic 
instrumentation system has been proven 
as a fieJd service facility developed well 
beyond the laboratory and experimental 
stage. 

Discussion 

John E. Voorhees (Battelle Memorial 
Institute}: What was the final /-number of 
this combined lens system? 

Mr. Bauer: We feel that the aperture 
value of the system is slightly less than 
that of the objective lens used. With the 
present //1. 5 objective, the practical 
aperture value of the system is about 
//I- 7. 

Mr. Voorhees: Wasn't the focal length 
changed? 

Mr. Bauer: Yes. The objective lens is 
racked forward to focus on the virtual 



image of the field lens. In order to bring 
the system to focus at infinity, the objective 
is set at a focal length of near 28 \ mm 
resulting in system focal length of about 
3f mm. 

William E. Cowles (General Electric Co.): 
Have you tried projecting the film records 
obtained from the optical system onto a 
hemispherical screen? 

Mr. Bauer: We have done a very limited 
amount of work using that technique. At 
one time, it was considered as a possible 
aid for photograrnmetric analysis of the 
film records. However, suitable tech- 
niques for data reduction of the film 
records have now been established as 
routine, using normal flat projection 
devices, such as the Eastman Recordak 
or our own Douglas Iconolog film reader. 

Mr. Cowles: That projection technique 
has been tried with another type of lens. 
I don't know if the characteristics of that 
lens and yours are the same. 

Mr. Bauer: I believe you may be re- 
ferring to the Jam Handy lens. At one 
time I had an opportunity to look at 
photographs of the assembled lens. Un- 
fortunately, pertinent schematics, reports, 
etc., were not available for study; there- 
fore I cannot compare the characteristics 
of that lens and the lens system described 
in this paper. 

E. P. Martz, Jr. (Holloman Air Force 
Base}: You note that experiments were 
made with filters. Does the use of filters 
reduce the advantage of using color film? 

Mr. Bauer: No. Lateral chromatism 
in this system is, of course, very apparent 
when using either color or black-and- 
white film. With color film, this effect 
works to our benefit ; with black-and-white 
film, the effect is very much a disadvantage. 
For example, consider a narrow, white 
post near the extreme edge of the field. 
On color film, this post will be beautifully 
fringed in colors and readily detected due 
to the color contrast. However, on black- 
and-white film, the white post will defi- 
nitely be obscured in the fuzzy gray 
images resulting from chromatic differences 
of magnification. 

Mr. Martz: The color filters you used 
were they monochromatic? 

Mr. Bauer: No. We used a Wrattcn 2B 
with Daylight Kodachrome, and Wratten 
No. 85 with Type A. The purpose of these 



Bauer and Blake: Wide-Angle Optics for High-Speed 



filters is to take advantage of the principle 
of using slightly less chromatic light while 
retaining color contrast. For our present 
work, this contrast is desirable. 

Mr. Martz: Have you made resolving 
power tests? If so, what order of magni- 
tude of resolution, in lines per millimeter, 
resulted? 

Mr. Bauer: We have made tests using 
Kodachrome film which has a resolving 
power of close to 60 lines per millimeter. 
The tests were conducted under conditions 
as near ideal as practical with a test set-up. 
Under these controlled conditions, near 
maximum resolution was obtained in the 
paraxial region. However, due to astig- 
matism and lateral chromatism. tangential 
resolving power dropped off to about 15 
and radial resolving power to approxi- 
mately 40 lines per millimeter at the 
extreme edge of the field. 

Anon: Would you care to make some 
comments on plastics for wide-aperture 
optics? 

Mr. Bauer: I can only give you informa- 
tion pertaining to our particular field. 
There is a very limited choice in index 
of refraction for optical plastics. Plexiglas 
is about 1.49. Should we attempt any 
corrections of the field lens, some glass 
will have to be used. For our present 
uncorrected field lens, we would realize 
only a few benefits from a glass version. 
Fluoride coatings on glass can be made 
more durable by baking, crazing would be 
eliminated, and the surface light scattering 
effect [haze] would be reduced. All of 
these factors were considered but the 
relatively low costs of plastic optics were 
very attractive. We have had some 
trouble with plastic crazing at high 
altitudes; however, it has not seriously 
interfered with our work. 

Winston O. Johnson (E. I. du Pont de 
Nemours & Co.): Does the haze problem 
apply when you have a fluoride coating? 

Mr. Bauer: We believe that the haze 
problem is reduced by coating, but the 



benefit provided by a fluoride coating is 
an increase in light transmission. At the 
present time, the coating cannot be baked 
on and is therefore soft, requiring extra 
care in handling. 

Anon: Can these lenses be used for short 
distances? 

Mr. Bauer: If the extreme depth of 
field does not encompass close objects 
when focused at infinity, this system can be 
focused to shorter ranges. 

Anon: What I am thinking of is a test 
chamber which is no more than 10 feet 
across. We would like to see everything 
in that chamber in focus. 

Mr. Bauer: I believe that the extreme 
depth of field of this system might be used 
to your advantage. 

Anon: Is it available commercially? 

Mr. Bauer: Yes. Wollensak Optical 
Company or Douglas Aircraft Company 
can supply the necessary equipment. 

B. J. Brettler (Edgerton, Germeshausen & 
Grier) : I wonder if you would comment a 
little on distortion calibration. 

Mr. Bauer: We place an array of targets, 
ten degrees apart, in a semicircle of 30- 
foot radius. Over the center of the circle 
is placed the node of admission of the 
optical system. With the targets and 
cameras carefully located, pictures are 
taken, and a distortion curve is determined 
from measurements made on the picture. 
Many of our optical systems were checked 
this way. We found that there was only 
a very slight difference between lenses; 
therefore, a standard distortion curve 
could be made. The optics of all the 
different types of cameras used were also 
rotated about the optical axis, and here 
again, the differences were negligible. 
From the distortion curves determined 
from the film records, we made an overlay 
grid to be used on a Recordak viewer or 
our own Iconolog. With this system, we 
are reading film records to well within plus 
or minus a degree, up to 50 degrees off 
axis. 



48 



January 1953 Journal of the SMPTE Vol. 60 



New Automatic Film-Threading 
Motion Picture Camera 

By G. J. BADGLEY and W. R. FRASER 



The new automatic film-threading motion picture camera designed and built 
by G. J. Badgley at the Naval Photographic Center provides: (1 ) a 16mm mo- 
tion picture camera that can be easily loaded, quickly threaded and operated 
under adverse conditions normally encountered by naval photographers; 
and (2) a motion picture camera that can be used for recording of radar and 
television images appearing upon cathode-ray tubes. 



JL HE modern professional motion pic- 
ture camera is a marvel of engineer- 
ing design and construction. Without 
tolerances that are, in some instances, 
measured in ten thousandths of an inch, 
it would not be possible to achieve the 
high degree of technical perfection in 
cinematography that we know today. A 
"must," and rightfully so, has been 
"rock-steady" motion pictures with ac- 
companying near-perfect registration. 
The loading and threading of film in such 
a camera, however, have been secondary 
in importance and considered to be 
necessary evils that could be minimized 
by employment of skilled and experi- 
enced motion picture cameramen. 

Such a solution, however, is not so 
easy where the Armed Forces are con- 
Presented on October 19, 1951, at the 
Society's Convention at Hollywood, by 
G. J. Badgley, who read the paper, and W. 
R. Fraser, Research and Development 
Dept., U.S. Naval Photographic Center, 
Naval Air Sta., Anacostia 20, Washington, 
D.C. 



cerned. A Navy Photographer, for ex- 
ample, may not touch a motion picture 
camera for days, weeks or even months at 
a time. He must be a "Jack of all 
trades." One day he may be an aerial 
photographer, another, a processing man, 
a still photographer, a color man or any 
one of a dozen or more occupations fall- 
ing under the general heading of "Naval 
Photographer." Being unable to con- 
centrate on any one field of photography, 
let alone on any one motion picture cam- 
era, he is interested primarily in a camera 
that is easy to load and operate under a 
great variety of conditions. Rarely will 
photography be conducted under the 
ideal conditions found on a sound stage , 
more likely the scene will be "shot" 
under enemy fire, in an aircraft or on 
board a ship at sea. Throw in the 
weather element both hot and cold 
and it becomes painfully apparent that 
our navy cinematographer has a few 
things on his mind other than his camera. 
It is well known that the amount of 
footage exposed varies inversely with the 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 1. A general view of the Badgley "Automatic Threading" motion picture 
camera. The 25-mm //0.7 Polaroid and 55-mm Bausch & Lomb high-speed lenses, 
turret locking ring, camera drive motor, and 400-ft film magazine are in position. 



difficulties encountered, and conse- 
quently it is important to insure that the 
camera is not another offender. Invari- 
ably, when environmental difficulties are 
encountered, the second roll of film re- 
mains unexposed. The reason, obvi- 
ously enough, is due generally to the 
laborious procedure required to thread 
the film in the camera. With this in 
mind, it was decided to attempt develop- 
ment of a motion picture camera that 
would have an automatic threading fea- 



ture, be usable in all kinds of weather by 
relatively untrained personnel and be 
practically foolproof in operation. 

From a study of both past and present 
cameras, it was found that while numer- 
ous improvements have been made in the 
camera body proper, very few effective 
improvements have been made in the 
magazine or film carrier. Some at- 
tempts have been undertaken to mini- 
mize the difficulties encountered in put- 
ting film into a camera and in shortening 



50 



January 1953 Journal of the SMPTE Vol. 60 






the time factor of threading and setting 
the loops. For instance, a series of cam- 
eras has been developed in which a large 
part of the camera mechanism the 
pulldown claws, the aperture and the 
film gate is incorporated in the maga- 
zine or film box. This method, however, 
results in an overly expensive magazine 
of which there must be several for each 
camera. Also, the fitting of these maga- 
zines to the camera requires a high order 
of precision which cannot be retained dur- 
ing the normal life of the camera due to 
wear and tear incident to service use. 

Other cameras developed were limited 
to magazines with a film capacity of 50 to 
100 ft with the necessary loops and 
sprockets incorporated within the maga- 
zine. However, this system was accom- 
panied by troubles arising from installa- 
tion of the necessary couplings and 
drives within the main body of the cam- 
era. Another system utilized feed 
sprockets incorporated within the maga- 
zines, but the accumulation of dust and 
dirt that could not be removed without 
completely dismantling the assembly was 
a source of trouble. In general, these 
camera types left a great deal to be de- 
sired in reliability, simplicity and ease of 
operation and maintenance. In sum- 
mation, the problem was reduced to the 
following : 

1 . To design and develop an automatic 
threading camera wherein the film will 
be threaded during the process of attach- 
ing the magazine to the camera. This 
also included the requirement that 
loaded magazines could be removed 
from the camera at a moment's notice 
and then re-used in the camera at any 
time thereafter; and 

2. To divorce all moving mechanisms 
(except the feed and reroll spindles) and 
their accompanying drives from the film 
box or magazine. 

The Experimental Model 16mm Camera 

In the process of developing a camera 
that would comply with the aforemen- 
tioned performance requirements, a 



16mm experimental model camera was 
designed, built and tested (Fig. 1). The 
camera weighs 1 5 Ib and has been used as 
the vehicle to prove the feasibility and 
practicability of the numerous features 
that it incorporates. As indicated by 
the title of this paper, the major feature 
permits automatic threading of the 
camera regardless of the film capacity of 
the magazine. Other features include 
unique magazine light-traps, high-speed 
film pulldown movement for television 
and radar scope recording, a special 
stabilized shutter to minimize "shutter 
bar," constant-speed main drive shaft to 
insure smoother camera operation, turret 
locking ring as an antivibration measure, 
quickly removable electrical assemblies, 
and special features to facilitate tele- 
vision and radar scope recording. 

By selecting certain features of the ex- 
perimental camera, it will be possible to 
develop special-purpose cameras includ- 
ing: (1) a combat version, also suitable 
as an amateur camera ; (2) a radar and/ 
or television recording camera that is 
portable and suitable for use in aircraft; 
and (3) professional versions, both 16mm 
and 35mm with reverse take-up and 
other refinements. 

In designing the camera, precision 
workmanship has been used where neces- 
sary. However, precisely fitting parts 
have been avoided where not required, as 
experience has shown that unnecessarily 
snug fits are detrimental especially 
under extreme high- and low-tempera- 
ture conditions and also in the presence 
of fine particles of dust or sand. In 
addition, all precisely related parts, 
wherever possible, have been constructed 
as integral units. Provision has been 
made to permit simple adjustments of 
moving parts that are subject to normal 
wear. 

Magazine 

The principal feature of the camera is 
automatic threading regardless of the film 
capacity of the magazine. During the 
act of attaching the loaded magazine to 



Badgley and Fraser: Automatic Threading Camera 



51 




Fig. 2. View showing position of the 400-ft film magazine 
just prior to engagement of film by sprockets. 




52 



Fig. 3. View of magazine seated in final locked position. 

Note film loops formed during process of "pushing" 

magazine forward into position. 

January 1953 Journal of the SMPTE Vol. 60 




Fig. 4. Internal view of 400-ft film magazine. The "nose piece," which contains 
the light-traps, is the heart of the magazine and it can be used in similar 
film magazines ranging from a capacity of 50 ft to 2000 ft or more. Note cover locking 
mechanism in lid. 



the camera, the two camera film sprock- 
ets are caused to move forward and also 
to rotate in such a manner that the 
sprocket teeth engage the film perfor- 
ations. As a result, two film loops of 
optimum length are always formed. 
The magazine becomes locked in position 
at the end of the throw, the entire opera- 
tion requiring less than two seconds. 
Figure 2 shows the magazine attached to 
the camera prior to being pushed for- 
ward while Fig. 3 shows the magazine 
locked in the final position. By means of 
this simple procedure, the magazine- 
changing problem is greatly reduced and 
the cameraman, although forced to use 
heavy gloves during cold weather condi- 
tions, will have a much better chance to 
complete his photographic assignment. 

In the conventional-type film maga- 
zine, the sides and base comprise the 
major part with the cover or lid being a 
circular, threaded plate. In the experi- 
mental magazine, however, the sides are 
part of the cover, thus leaving a clear, 



flat plate with film spindles and guide 
rollers (Fig. 4). This feature facilitates 
loading since there are no sides to inter- 
fere with placement of the film on the 
spindles. The cover lock supports the 
free end of the film spindles, thereby pro- 
viding additional mechanical support to 
these members. The cover slips on over 
the base plate and may be positively 
secured in position by a simple one-twist 
operation. 

Loading of the magazine has been sim- 
plified by the design of unique light- 
traps that are symmetrical and inter- 
changeable and which may be readily 
disassembled for cleaning and then re- 
assembled all without the use of tools. 
These light-traps have been so designed 
that the film during loading in the mag- 
azine must follow the only path available 
through the light-traps. 

One pair of side guides for the film 
aperture is in the camera and is fixed 
relative to the pulldown pins. The other 
pair, which is in the magazine, is mov- 



Badgley and Fraser: Automatic Threading Camera 



53 



able and spring-loaded. However, the 
operating spring for this movable pair in 
the magazine is in the camera and does 
not close the guide on the film until the 
magazine is fully seated and the back 
pressure plate has seated the film 
against the aperture plate. 

Design of the magazine reverse take-up 
is unusual in that a nonrachet-type take- 
up that completely disengages the feed- 
roll spindle is employed. Operation is 
dictated by the direction of rotation of 
the driving gears which couple or un- 
couple with the particular film spindle 
concerned. 

High-Speed Pulldown 

Besides automatic loading, the major 
innovations in this camera have been 
dictated by the high degree of precision 
required for television recording. This 
camera has a high-speed pulldown move- 
ment which permits use of the 288 shut- 
ter opening for kinerecording at 24 
frames/sec. An ample safety margin of 
operating coverage by the shutter during 
movement of the film is provided since 
there is approximately a 10 shutter 
coverage both before and after exposure. 

Stabilized Shutter 

One of the major problems in the con- 
struction of a television recording camera 
is the design of a shutter movement that 
is completely free of backlash. This was 
accomplished by tying all related parts to 
a common shaft; that is, the advance 
cam, motor drive and shutter are all on 
the main shaft. There are no loose 
couplings that can produce objectionable 
backlash. 

In order to achieve further stability, a 
weighted flywheel rim is loosely mounted 
around the periphery of the shutter 
member and runs between the pole 
pieces of a magnet. In this way the 
mechanical dampening action of the 
slipping flywheel and the electrical drag 
or braking action of the hysteresis current 
generated in the shutter rim are both 
utilized. 



Constant-Speed Main Drive Shaft 

The main drive shaft of the camera 
rotates at a constant speed at all times 
during the entire cycle of operation. 
This is in contrast to several motion pic- 
ture cameras that utilize an acceleration 
mechanism to speed up the pulldown 
cam during the film-transfer portion of 
the cycle, with a corresponding slow- 
down of the pulldown cam during the 
exposure portion of the cycle. This in- 
termittent velocity is not conducive to 
steady shutter travel, and as a result the 
shutter must be driven by other means 
such as a separate synchronous motor, 
for example. This method, however, 
means added weight and the mounting 
of the motor precludes the use of turret 
lenses. The added weight, interlocking 
mechanisms and extra wiring all detract 
from portability which is so important in 
airborne operations. 

Turret Locking Ring 

When cameras are used in reciprocat- 
ing engine type aircraft, a certain amount 
of vibration will be transmitted to the 
camera and as a result local vibrations 
may be generated between the various 
parts of the camera. These vibrations 
may be of sufficient severity to affect the 
photographic image, and to prevent this, 
use is made of a large threaded ring that 
may be screwed up against the outer 
periphery of the turret. This ring, which 
resembles the ring gear of an automobile 
flywheel, is rotated by a manually oper- 
ated pinion gear, which action effectively 
locks the lens turret against the front 
plate of the camera (Fig. 5). Also, an 
indexing pin is employed to accurately 
position the turret and to center the 
lenses. 

Electrical Components 

In the design of the camera electrical 
circuits, protection against weather and 
mechanical damage, and simplicity with 
reliability were the prime considerations. 
To achieve these ends, all wiring is in- 



54 



January 1953 Journal of the SMPTE Vol. 60 







Fig. 5. View showing focusing viewfinder in position. 



corporated within the camera body. 
Attachment of the motor to the camera 
automatically completes the electrical 
connections through the built-in plugs in 
the motor and camera base. This fea- 
ture provides firm, self-aligning, wear- 
compensating mountings for quick at- 
tachment and removal of the motor as 
well as for extreme simplicity 

The power-feed receptacle, reverse 
switch, buckle switch, ON-OFF switch, 
and red light warning assembly are built 
as separate units. These electrical 
assemblies, which are commercially 
available, may be quickly removed from 
the camera for replacement or mainte- 
nance without disturbing or removing 
the soldered connections or any parts of 
the camera. The buckle switch stops 
the camera and a warning red light 
flashes when there is loss of loop. Fur- 
thermore, when the magazine is im- 
properly positioned the camera will not 
start. 

The pendant hand switch and power- 
feed cable comprise the external wiring 



of the camera. Most cameras have the 
ON-OFF switch on the body of the cam- 
era, but in order to eliminate disturbance 
of the camera by physical contact of the 
operator during shooting, the switch has 
been placed on the end of a pendant 
cord. More important, however, is the 
fact that the pendant-switch method per- 
mits use of relays, timing mechanisms 
and other remote-control systems as de- 
sired. 

Television Recording 

As reported in numerous articles on 
the subject of kinerecording, elimination 
of the phenomenon known as "banding" 
or "shutter bar" has been extremely diffi- 
cult to accomplish either by mechanical 
or electronic means. It is well known 
that a motion picture camera equipped 
with a 288 open shutter and operated at 
the synchronous speed of 24 frames/sec 
is used to record standard 30 frame/sec 
commercial telecasts. However, numer- 
ous and apparently insignificant factors 
including shutter flutter, mechanical 



Badgley and Fraser: Automatic Threading Camera 



55 



backlash, film creepage, distance of the 
shutter blade from the focal plane, the 
//stop of the lens, humidity, temperature, 
variations in synchronism between the 
camera drive motor and the television 
frequency, and frequency variations will 
produce a certain amount of shutter bar. 
Normally, two shutter bars will appear 
when the kinerecording is projected. 
However, by proper synchronization of 
the camera with the television scanning 
frequency, one of the bars can be phased 
out of the picture area leaving the re- 
maining bar in the center of alternate 
film frames. The size or width of this 
bar depends upon the effectiveness of the 
"video splice" 1 which in turn is depend- 
ent upon the forementioned factors of 
shutter flutter, etc. In general, the 
practice in commercial television studios 
has been to introduce all the refinements 
and tricks of the trade possible and then, 
by trial and error, adjust the recording 
camera shutter opening by increments of 
0.0001 in. or less until the optimum con- 
dition has been determined. 

From the point of view of the Navy, 
this long drawn out calibration procedure 
is not desirable and cannot be tolerated 
since the Navy camera will be used at 
many locations in connection with vari- 
ous television systems. Consequently 
the Navy's kinerecording equipment 
must be portable, compact, lightweight, 
rugged and versatile. Furthermore, 
quick and easy adjustment of the camera 
in the field to meet changing electrical 
and electronic conditions must be within 
the capabilities of the cameraman. 

In order to handle the various types of 
kinerecording problems both airborne 
and underwater that are normally en- 
countered in the Navy, several additional 
features were deemed necessary. It was 
considered desirable for the cameraman 
to be able to view the image on the kine- 
scope through the high-power viewfinder 
(Fig. 5) during camera operation in 
order to check for the presence of unde- 
sirable banding or incorrect phasing. 



This can be done by inserting a prism 
into the focal plane of the photographic 
aperture by manipulation of a small 
spring-loaded lever located below the 
focusing tube (Fig. 5). The cameraman 
may then view the translucent image 
through the film while kinerecording is 
in progress, or view the image as re- 
flected by a fine, ground-glass prism 
when the camera is operating without 
film. In the first instance, the film acts 
as a ground glass and the operator actu- 
ally sees the image that is being exposed 
and recorded on the film. Viewing of 
this image during exposure, however, 
will not result in fogging of the film. In 
the event that two shutter bars are seen, 
one of the bars can be "phased out" 
immediately by rotation of a control 
knob that will vary the electrical phase 
position of the armature of the camera 
synchronous motor with the camera 
movement. The width of the remaining 
band may be quickly reduced to a negli- 
gible value by use of a screwdriver ad- 
justment of a fine worm gear that 
changes the shutter opening in micro- 
scopic increments. 

The object distances normally em- 
ployed in kinerecording are unusually 
short in the neighborhood of 15 to 18 
in., and this condition puts a premium on 
obtaining a sharp focus. A 25 X view- 
finder was therefore incorporated in the 
camera to facilitate quick and accurate 
focusing. 

FCC-approved CBS "Field Sequen- 
tial" color telecasts can be recorded in 
color provided a high-aperture lens is 
employed. 4 Furthermore, use of the 
300 open shutter will permit recording 
of five sequential color fields with film 
pulldown occurring during the sixth or 
"blue" field. With the use of this tech- 
nique, more than 83% of the color in- 
formation will be recorded as compared 
to 50% recorded when the 180 shutter 
is employed. RCA color can also be re- 
corded on color film using a high-speed 
lens and the 288 open shutter. 



56 



January 1953 Journal of the SMPTE Vol. 60 






Radar Scope Recording 

Plan Position Indicator (PPI) radar 
scope recording, 2 - 4 unlike television, does 
not require synchronization although, 
quite naturally, a new set of problems is 
introduced. High-speed lenses with 
apertures of //I or faster are generally 
required to record satisfactorily the per- 
sistent yellow or green component of the 
images appearing on PPI radar scopes. 
High-speed lenses have short back-focus 
distances which make it necessary to 
locate the plane of the shutter as close to 
the focal plane as is physically possible. 

In the experimental camera, the 288 
open shutter used for kinerecording will 
be replaced by a shutter with a fixed 
maximum opening of approximately 
310 for radar scope recording purposes. 
This unusually large shutter opening will 
permit, in some instances, the use of 
standard, high-quality //1 .4 cine lenses. 
The high-speed lenses on the camera 
(//0.7 and //0.9) are approximately 
three inches in diameter which accounts 
for the rather large diameter of the lens 
turret. It may be mentioned in passing 
that in the design of high-aperture 
lenses, 5 barrel distortion, astigmatism 
and other abberrations are tolerated in 
order to gain the extra speed that is so 
essential. These lenses, in addition to 
having relatively poor optical qualities, 
are very expensive and consequently are 
employed only when the radar scope 
light levels are so low that their use be- 
comes necessary. 

The importance of radar scope cine- 
matography in the Armed Forces is in- 
creasing and with this new camera it will 
be possible to make scope recordings that 
previously were not feasible. In the 



field of radar training-film production, it 
has been necessary in the past to utilize 
animation techniques almost exclusively. 
Actual photography of the radar scope 
will create the realism that is so essential. 
For added realism, the film may be 
projected through a green or amber filter 
to duplicate more closely the appearance 
of the radar scope. This realistic effect 
may also be obtained by projecting the 
film upon a screen treated with the new 
green and amber colored fluorescing 
materials. 

Conclusion 

The experimental camera is geared 
primarily to the solution of television and 
PPI radar recording problems. The 
heart of the camera, that is, the rapid 
self-threading feature that permits a 
magazine to be changed in less than ten 
seconds, would be highly desirable in 
professional, recording, amateur and 
combat motion picture cameras. 

References 

1. F. N. Gillette, "The picture splice as a 
problem of video recording," Jour. 
SMPE, 53: 242-255, Sept. 1949. 

2. R. G. Babish, "Radar scope photog- 
raphy," Jour. SMPE, 48: 454-472, May 
1947. 

3. F. X. Clasby and R. A. Koch, "War- 
time naval photography of the elec- 
tronic image," Jour. SMPE, 50: 189- 
198, Mar. 1948. 

4. W. R. Fraser and G. J. Badgley, "Motion 
picture color photography of color tele- 
vision images," Jour. SMPTE, 54: 735 
744, June 1950. 

5. E. K. Kaprelian, "Objective lenses of 
//I aperture and greater," Jour. SMPE, 
53: 86-99, July 1949. 



Badgley and Fraser: Automatic Threading Camera 



57 



Animation Stand of New Design 



By E. H. BOWLDS 



An animation stand engineered to meet all requirements of flat-bed stop- 
motion animation for 16mm and 35mm film is described. It is designed to 
combine ease of operation with facilities for the most intricate effects shots 
and special techniques. A unique peg and platen system allows larger field 
sizes than hitherto available on similar equipment. The use of ball bearings 
at all friction points eliminates power-driven mechanisms and simplifies 
maintenance. 



I 



.N RECENT YEARS, the widespread use of 
animation for training films and other 
nontheatrical productions has stimulated 
the demand for reliable, compact, easy- 
to-operate animation equipment which 
could be economically manufactured so 
as to be within the reach of the average 
independent producer. Therefore, a 
study was made to plan and engineer an 
animation stand which would fulfill the 
needs of the greatest possible number of 
users. Consultation with individual 
operators, special-effects men and ani- 
mators, together with our own extensive 
experience in the animation field, pro- 
duced a long list of features which should 
be incorporated into such an animation 
stand. To combine the most desirable 
of these features into one unit at eco- 
nomical cost presented a challenging 
engineering problem. 

Presented on October 7, 1952, at the 
Society's Convention at Washington, D.C., 
by Benjamin Berg for the author, E. H. 
Bowlds, E. H. Bowlds Engineering Co., 
1507 N. Kingsley, Los Angeles, Calif. 



Over two years went into laying the 
groundwork, revising plans and eliminat- 
ing unnecessary components, before blue- 
prints were ready to present. Church- 
ill-Wexler Film Productions of Los 
Angeles were the first to feel that these 
plans suited their financial and mechan- 
ical requirements, and so they sponsored 
construction of the first model (Fig. 1). 

Requirements 

This pilot animation stand was to pro- 
vide the following features : 

1. A maximum vertical travel from a 
close range of 3j to a 16 field (on the 
basis of the "Acme" system). 

2. A movable table with a horizontal 
range of 12 in. east or west of the optical 
center, and a 4^-in. movement north or 
south from center. 

3. A pantograph mechanism with an 
adjustable pointer on an adjacent field 
for the purpose of quickly plotting diag- 
onal or irregular movements. 

4. A recessed box for back-lighting in 
the center of the table. 

5. 360 rotation of the camera. 



58 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 1. Assembled animation stand with 16mm Cine Special Camera 
mounted on camera arm. 



. H. Bowlds: Animation Stand of New Design 



6. A peg-bar and platen system which 
would allow rapid conversion from a 
standard 12-field to an oversize 16-field. 

7. The stand was to accommodate 
either 16mm or 35mm camera by inter- 
changeable mounts on the vertical car- 
riage. 

Design and Construction 

The basic construction consists of 
three components: the base which sup- 
ports the entire stand, the table base, 
and the upright for vertical camera 
movement. These parts are cast-iron, 
normalized before machining to insure 
an adherence to close dimensional toler- 
ances. 

The base, with flanges, is 24 in. square 
and has a leveling screw in each corner. 
Secured to this is the table base with 
machined top and machined rails to 
accommodate the north-south movement 
unit. In back, this table is recessed to 
allow for the upright, which consists of a 
heavy tee casting, machined to accom- 
modate ball bearings for a friction-free 
vertical carriage. 

The various components are built on 
this skeleton structure (Fig. 2). The 
table is made of seasoned birch. It is 
secured to heavy steel rails that ride on 
adjustable ball bearings for the east-west 
movement. These bearings are under 
tension to eliminate any play in the 
table travel and to assure ease of move- 
ment. A rule is set into the front of the 
table for easily visible east-west table 
calibrations. 

The north-south movement is con- 
structed in the same manner; however, 
this is activated by a lead screw with vee- 
threads and a bronze hex nut cut in seg- 
ments, with spring take-up to avoid any 
backlash. It is calibrated by a tape rule 
running over a drum behind the hand 
control, to the right of the operator. 

Next, a unique peg-bar system was 
devised to allow for interchangeable 
fields. Heavy shafts were mounted in 
ball bearings at each end of the table. 
On each shaft are two drums of 12-in. 



circumference. These are normally set 
for a 12-field but may be shifted to 
accommodate larger fields. Over each 
set of drums is a spring-steel band on 
heavy tension which can be released by a 
snap device. Standard peg bars, 3 
fields in length, are secured to these 
steel bands, and are activated by hand- 
wheels. The wheel to the right of the 
operator moves the bottom pegs east or 
west. 

A similar wheel to the left moves the 
top pegs. The wheels are calibrated so 
that top and bottom pegs may be moved 
at different speeds. The peg bars them- 
selves may be removed in a few seconds 
leaving the sponge-rubber table top 
clear; or the bars may be spread apart 
with the adjustable drums to give an 
oversized field. 

A manual platen with a handbar was 
installed on this model. This platen is 
of the usual type with water-white glass 
and a spring device which allows it to 
remain open or closed. The pressure 
can be adjusted by placing more tension 
on the springs. The platen may be re- 
moved by loosening one screw. An 
oversize platen may then be put in its 
place. This change-over takes but a 
few seconds and provides a feature which 
has not hitherto been available on 
animation stands. 

Another important feature on the 
table is the pantograph. In practice, 
the use of the pantograph has been 
found to be an enormous time saver. 
The operator draws the path of his de- 
sired camera movement on a piece of 
paper equivalent to the field size he is 
shooting. This path may be a straight 
line, a curved line or any complex shape. 
The needle of the pantograph, which is 
stationary, indicates at all times the rela- 
tion of the optical center of the camera to 
the table. Thus by moving the table 
with the pantograph needle as a reference 
point, the operator can describe any 
move he wants, no matter how intricate. 
This system avoids detailed and lengthy 
calibrations and considerably speeds up 



60 



January 1953 Journal of the SMPTE Vol. 60 




Fig. 2. Table assembly, showing platen, pantagraph, peg bars 
and control wheels. 



shooting time. It may be folded out of 
the way when not in use or brought into 
position on the righthand side of the 
table. The pointer on the pantograph 
is also adjustable. 

The camera carriage is engineered on 
a ball-bearing system similar to that used 
on the table. It follows the precision- 
machined vertical track under heavy 
tension to assure the same accuracy of 
movement. The carriage is activated 
on a steel cable arranged for pulley ac- 
tion. The operator controls this mecha- 
nism by a handwheel to his left. Com- 
bined with this handwheel is a brake or 
clutch to hold the vertical unit in any 
position, and to control the amount of 
camera weight transmitted to the hand- 
wheel. 

Vertical movements, as small as the 
width of the hair line on the vertical 



scale, are easily controlled by the oper- 
ator. This is desirable for making 
smooth vertical trucks. By releasing 
the clutch, the entire vertical carriage 
can be moved up or down very rapidly 
and the fact that ball bearings have 
been used at all friction points, both 
rotary and sliding, has made it almost 
effortless to make any of the intricate 
movements required of this equipment. 

Counterweights were considered to 
relieve the pressure on the manual con- 
trol of the vertical movement. On the 
pilot model, this unit functioned so 
easily that no counterweights were 
necessary. However, they may be 
added, should a heavier camera demand 
it. 

Next on the vertical unit (Fig. 3), 
placed on a pivotal point over the optical 
axis, is a ring which allows 360 rotation 



E. H. Bowlds: Animation Stand of New Design 



of the camera. The ring rotates on a 7- 
in. bearing which can be locked at any 
degree. Calibrations are on the edge 
directly visible to the operator. A 4j- 
in. aperture in the center will accommo- 
date a large camera lens or lens bell for 
focusing. 

The camera mount is secured directly 
to this ring with wing nuts and registry 
pins. Both rotation disc and camera 
mount have adjusting screws for vertical 
drift and field rotation. 

The camera assembly on the pilot 
model includes an Eastman Kodak 16- 
mm Cine Special camera, a Richardson 
animation motor, and a fading mecha- 
nism coupled with the camera's dissolv- 
ing shutter. Two important consider- 
ations influenced the design of the shutter 
control and coupling device. First, 
there could be no backlash between the 
calibrating arm and the camera's shutter. 
Secondly, the position of the control it- 
self is important. Because fades and 
dissolves take up a large part of the 
operator's time in animation shooting, 
these controls should be immediately 
accessible and visible. Both these con- 
siderations were satisfied by the inclusion 
of a horizontal quadrant directly in 
front of the operator's line of sight and 
conveniently in reach from a sitting posi- 
tion. This quadrant contains the shut- 
ter calibrations from to 100% expanded 
over a 6-in. scale. The pivoted arm 
which is coupled to the camera's shutter 
mechanism holds a scribed plastic win- 
dow directly over the chart on the quad- 
rant. Fades of any length are accom- 
plished by moving this window hori- 
zontally over the stationary calibrated 
chart. 

A synchronous Richardson Stop- 
Motion Motor was selected for the cam- 
era drive. This motor is noiseless in 
operation and the design insures uniform 
exposure on continuous run as well as 
single frames, forward or reverse. It is 
also easily interchanged between 16mm 
or 35mm cameras and incorporates a 
large-size frame counter. The motor 



can be activated by a foot switch or a 
hand button. The on-off, forward- 
reverse switches are mounted in a small 
control box which may be placed within 
convenient reach of the operator. 

A bell-shaped ring is secured to the 
focusing scale of the Ektar Lens in use on 
the pilot model. This bell is calibrated 
in quarter-fields to correspond with the 
vertical movement of the camera mount. 
The bell attachment expands the focus 
calibrations to over double the radius of 
the original lens focus ring. To focus 
the lens, the calibrated bell is rotated 
behind a scribed Lucite bar. 

The lens aperture is adjusted by mov- 
ing a 3-in. arm which is secured to the 
//stop ring of the Ektar lens. 

This entire camera assembly, including 
the motor unit, may be removed through 
the use of two wing nuts. Registry pins 
insure accurate registration of the optical 
system with the mount's rotating axis. 
In this way, a very rapid change-over 
can be made from a 16mm to a 35mm 
camera. 

Finally, a light box 9 X 6 X 12 in. j 
was incorporated into the table (Fig. 4) 
to allow photography of pencil tests or 
transparencies. The light source may be 
flourescents, or a grid of closely spaced 
neon tubing. 

The overall dimensions of the stand 
are as follows : 

Floor to top of upright, 8 ft ; 

Width of table, east-west, 3j ft; 

Front of table to rear of assembly, 3 ft ; 

Height of table, 33 in. ; 

Face of upright to optical center, 14 in.; 

Maximum vertical camera travel, 45 in. ; 

Maximum east-west table movement, 12 

in. from center; and 
Maximum north-south table movement, 

4j in. from center. 

All exposed surfaces of the equipment 
are black anodized to avoid light reflec- 
tions getting back to the optical system. 

Lights to illuminate the platen are 
usually hung from the ceiling or the wall. 
One 500-w Baby Keg light of the Mole- 



January 1953 Journal of the SMPTE VoL 60 







Fig. 3. Camera mount, fading and focusing mechanisms and 
stop-action motor. 




Fig. 4. Light well, directly underneath the platen. 
E. H. Bowlds: Animation Stand of New Design 



63 



Richardson or Bardwell McAlister type, 
on each side of the stand, has been found 
to provide an evenly illuminated field 
whose intensity can be conveniently con- 
trolled. Polaroid filters over the light 
source and the lens help eliminate flashes 
from transparent overlays and other un- 
desirable reflections. 

A few words are in order to illustrate 
briefly the simplicity of putting the 
stand into operation. 

The leveling bolts in the base are first 
adjusted to compensate for any irregu- 
larities in the floor and to level the stand 
itself. The camera unit is then placed 
on the registry pins, the wing nuts tight- 
ened, and two adjustments are made. 
First the camera aperture is projected to 
an arbitrary rectangle on the table and 
the camera is rotated to determine if the 
optical axis is exactly perpendicular to 
the plane of rotation. Three small set- 
screws, located at the base of the camera 
mount, allow for quick adjustment of the 
mount to insure rotation of the camera on 
exact optical center. 

The second camera adjustment is for 
possible drift during the vertical move- 
ment of the camera mount. If the axis 
of the lens is not exactly parallel to the 
line of vertical movement, the projected 
aperture image will drift either north, 
south, east or west during a move from a 
small to a large field. The amount of 
drift depends on the relationship of the 
lens axis to the vertical movement. The 
slightest deviation from the small toler- 
ances we allowed ourselves would intro- 
duce this drift. To compensate for 
possible errors or dimensional changes, 
four setscrews are located in the under- 
side of the camera mount. These 
screws may be adjusted to insure elimina- 
tion of any drift as the camera moves 
from the closest to the largest fields. It 
might be noted that in the pilot model 
there was no measurable drift, so the set- 
screws were left in neutral position. 

The peg bars and table bed may now 
be centered north, south, east and west 
by the same aperture projection tech- 



nique. The field size and focus calibra- 
tions are made and film tests are run to 
verify all adjustments and calibrations. 
Once these adjustments are made and 
the setscrews tightened, they need never 
be changed as long as the stand is not 
bodily moved from one place to another. 

Accessories 

Because of the simplified basic design, 
it is possible to add a number of acces- 
sories for situations requiring special 
equipment. The present hand-operated 
platen may be replaced with a mecha- 
nized platen movement. Air and hy- 
draulicly activated mechanisms were 
considered for a foot-operated platen. 
However, it was felt that these devices 
require careful maintenance to be 
trouble-free. So a new design was 
evolved which operates on a motor- 
driven cam. This should have the ad- 
vantage of being almost noiseless and will 
require little, if any, maintenance. An 
auxiliary set of flip pegs may be installed 
on the table, either north or south, to 
provide two additional stationary levels 
when both peg bars are being moved. 
Side pegs may also be added for such 
effects as panning long backgrounds, 
north and south through the field, 
although the ease of rotating the camera 
to 90 will take care of most of such re- 
quirements. 

Conclusion 

In operation, this simplified animation 
stand has proved its merits in timesaving. 
All controls may be conveniently reached 
from a sitting position. The stand has 
also proved equal to all types of intricate 
movements. The pantograph has cut 
shooting time considerably because all 
moves are simple to plot. The accurate 
calibrations within visual range give the 
operator a sense of assurance of his rela- 
tive positioning at all times. In the 
mechanism itself, the elimination of 
backlash or play relieves all mental 
calculations for such discrepancies. The 



64 



January 1953 Journal of the SMPTE Vol. 60 



absence of mechanized units has greatly 
simplified maintenance, and the com- 
pletely quiet functioning of all moving 
parts has helped increase the efficiency 
of the cameraman. Rapid and rhyth- 
mic operation has been encouraged be- 
cause all the controls are within mini- 
mum reach. 

We believe this equipment fulfills the 
goal we set out to achieve: a versatile 
animation stand, economical to manu- 
facture. 



Acknowledgment 

The author would like to acknowledge 
the sponsorship of this pilot model 
by Churchill-Wexler Film Productions 
whose staff, together with Lee R. Rich- 
ardson, of the Richardson Camera Co.. 
contributed through cooperation in tech- 
nical consultations. The author is also 
grateful for the suggestions and encour- 
agement of all those members of the 
animation industry who have shown in- 
terest in this new project. 



E. H. Bowlds: Animation Stand of New Design 



65 



Standards PH22.1, -.84, -.85 and -.92 
Positive-Negative Raw Stock Dimensions; 
16mm Projection Lamps; Enlargement Ratio 



ON JANUARY 8, 1953, the American Standards Association approved four new 
standards which are published on the following pages. 

PH22.1-1953, Dimensions for 35mm Motion Picture Film Alternate Standards 
for Either Positive or Negative Raw Stock. 

PH22. 84-1 953, Dimensions for Projection Lamps Medium Prefocus Ring Double- 
Contact Base-Up Type for 1 6mm and 8mm Motion Picture Projectors. 

PH22. 85-1 953, Dimensions for Projection Lamps Medium Prefocus Base-Down 
Type for 16mm and 8mm Motion Picture Projectors. 

PH22. 92-1 953, Enlargement Ratio for 16mm to 35mm Optical Printing. 

The first of the above standards was developed by the Film Dimensions Committee 
under the chairmanship of E. K. Carver and was published for trial and comment 
in the April 1949 and September 1951 Journals. It is the latter version which has 
now been given final approval. 

The next two standards were initiated by the Committee on 16mm and 8mm 
Motion Pictures, at that time chaired by H. J. Hood. After publication in the 
February 1951 Journal, and in addition to attention in the usual cinematographic 
standards channels, these standards were also reviewed by the ASA Sectional Com- 
mittee on Electric Lamp Bases and Holders, C81. This group proposed several 
modifications of PH22.84 which were subsequently accepted and are now incorpo- 
rated in the final standard. 

PH22.92 is a product of the Laboratory Practice Committee, J. G. Stott, Chairman. 
This was published as a proposal in the January 1952 Journal and was approved 
without change. H.K. 



66 January 1953 Journal of the SMPTE Vol.60 



American Standard 

Dimensions for 35mm Motion Picture Film *.?,.*. 
Alternate Standards wna-i-itu 
for Either Positive or Negative Raw Stock ' UDC 7783i77 l 




D 
90 /" 


^~ 

D 


- ^^^_ _^ 




Po 9 , 1 of 2 p.. 

JL 

B 

T 


A 


a 


LJ 

a. 




a 


a 
a 
:a 
a 
a R 

CD 

I H . - ' 


a_ 

n_ 




^p p 


n 


r 


n 










i 








T 




Dimensions 


Inches 


Millimeters 


A 
B 
C 
D 
E 
G 
1 
L* 
R 


1.377 d= 0.001 
0.1870 0.0005 
0.1100 =t 0.0004 
0.0730 0.0004 
0.079 it 0.002 
Not > 0.001 
0.999 dt 0.002 
18.700 db 0.015 
0.013 0.001 


34.980 db 0.025 
4.750 it 0.013 
2.794 it 0.01 
1.85 0.01 
2.01 0.05 
Not > 0.025 
25.37 0.05 
474.98 =t 0.38 
0.330 =t 0.025 


These dimensions and tolerances apply to the material immediately after 
cutting and perforating. 
* This dimension represents the length of any 100 consecutive perfora- 
tion intervals. 


Approved January 8, 1953, by the American Standards Association, Incorporated 
Sponsor: Society of Motion Picture and Television Engineers uirui tv r ,.i ru-.tr.ix,. 



Copyright 1953 by the American Standards Association. Incorporated 
70 Ent Forty.fifth Street, New York 17. N Y 



Price. : 



January 1953 Journal of the SMPTE Vol. 60 



67 



Page 2 of 2 pages 



Appendix 

(This Appendix is not a part of the American Standard Dimensions for 35mm Motion Pic- 
ture Film Alternate Standards for Either Positive or Negative Raw Stock, PH22. 1-1953.) 



The dimensions given in this standard represent the 
practice of film manufacturers in that the dimensions 
and tolerances are for film immediately after perfora- 
tion. The punches and dies themselves are made to 
tolerances considerably smaller than those given, but 
owing to the fact that film is a plastic material, the 
dimensions of the slit and perforated film never agree 
exactly with the dimensions of the punches and die:,. 
Shrinkage of the film, due to change in moisture con- 
tent or loss of residual solvents, invariably results in 
a change in these dimensions during the life of the 
film. This change is generally uniform throughout the 
roll. 

The uniformity of perforation is one of the most im- 
portant of the variables affecting steadiness of pro- 
jection. 

Variations in pitch from roll to roll are of little sig- 
nificance compared to variations from one sprocket 
hole to the next. Actually, it is the maximum variation 
from one sprocket hole to the next within any small 
group that is important. 

Perforations of this size and shape were first de- 



scribed in the Journal of the SMPE in 1932 by Dubray 
and Howell. In 1937, a subcommittee report reviewed 
the work to date. The main interest in the perforation 
at that time was in its use as a universal perforation 
for both positive and negative film. The perforation 
has been adopted as a standard at this time largely 
because it has a projection life comparable to that of 
the perforation used for ordinary cine positive film 
(American Standard Cutting and Perforating Dimen- 
sions for 35mm Motion Picture Positive Raw Stock, 
Z22.36-1947, or the latest revision thereof approved 
by the American Standards Association, Incor- 
porated), and the same overall dimensions as the 
perforations used in the negative film (American 
Standard Cutting and Perforating Dimensions for 
35mm Motion Picture Negative Raw Stock, Z22.34- 
1949, or the latest revision thereof approved by the 
American Standards Association, Incorporated). It 
should be particularly noted that although the pres- 
ent standard has the same overall dimensions as the 
older cine negative perforation, positioning pins or 
sprocket teeth made to fit this perforation exactly will 
injure the corners of the cine negative perforation. 



PH22.1-1953 



68 



January 1953 Journal of the SMPTE Vol. 60 



ft 



American Standard 

Dimensions for Projection Lamps 
Medium Prefocus Ring Double-Contact Base-Up Type 

for 16mm and 8mm Motion Picture Projectors 



ASA 

* f . v. 4. rtt. of. 
PH 22. 84- 195 3 



UOC 778.15:621 326.73 




+ 0.005 



ELECTRICAL CONTACTS 
SEE PAR 3 



THIS SIDE TOWARD 
CONDENSER LENS 



VENTILATING PORTS 
SEE PAR I 



0.094 0.005 R 



THESE THREE POINTS 
REGISTER AGAINST 
FIXED SURFACE IN 
LAMP HOLDER 



-CONDENSER SIDE 
OF LAMP 



-SOURCE CENTERED 
ON AXIS OF PRE- 
FOCUS RING WITHIN 
0.030, FRONT AND 
SIDE VIEWS 




1. Scope. The purpose of this standard is to 
establish, for the type of lamp shown, the di- 
mensions essential to Intel-changeability of 
lamps in projectors. It is not intended to pre- 
scribe either operating characteristics or de- 
tails of design such as the shape of the ven- 



ALL DIMENSIONS IN INCHES 



tilation ports or method of attachment of the 
prefocus ring to the base. 

2. Operating Position. Lamps of this type 
are intended to be burned with the axis in an 
essentially vertical position, and with the base 
at the top. 



Approved January 8, 1953, by the American Standards Association, Incorporated 
Sponsor: Society of Motion Picture and Television Engineers 



Price. 25 Ontt 



January 1953 Journal of the SMPTE A Vol. 60 



69 



Page 2 of 2 pages 

3. Electrical Contacts. The drawing indi- indicated when the lamp is inserted in a holder which 

cates the area which the electrical members of rotationally positions the lamp as shown in the end 

the lamp holder should contact. It is not in- view of the base - Condensing lenses, the mirror, and 

tended to dictate the shape of the terminals on ** mounts must ther ^ore be so located as to insure 

the lamp; however, they should not exceed ad * quate clearance between these P arts and the bulb 
boat-shaped areas 0.250 inch wide by 0.410 

inch long with the long axes parallel to the Note 2 . For med!um prefoc(JS base . down proiect i on 

plate on the flange. With lamps of this type, , amps/ see American Standard Dimensions for Pro- 

the prefocus ring is not an electrical contact. j ect i on tamps Medium Prefocus Base-Down Type for 

Note 1. These dimensions define the maximum ex- 16mm and 8mm Motion Picture Projectors, PH22.85- 

cursion of the bulb surfaces from the base axis toward 1953, or the latest revision thereof approved by the 

the condensing lenses and the mirror at the points American Standards Association, Incorporated. 



PH22.84-1953 



70 January 1953 Journal of the SMPTE Vol. 60 






American Standard 

Dimensions for Projection Lamps 
Medium Prefocus Base-Down Type 

for 16mm and 8mm Motion Picture Projectors 



K'l V. 5. ru Of. 

PH22.85-1953 



UOC 778.55-621 326 73 




THIS SIDE OF BULB 
TOWARD CONDENSER 
LENS 



FILAMENT 



TOP VIEW OF BASE 



I 0.725 MAX 

SEE J T-IOBULI 

NOTE n 0.850 MAX 
I T-12 BULB 




' 0.668 MAX"! 

T-.OBULB I SEE 
0.785 MAX [NOTE 1 

T-12 BULB I 



SOURCE CENTERED 
ON AXIS OF PRE- 
FOCUS BASE WITHIN 
0.030, FRONT AND 
SIDE VIEWS 



THIS SURFACE REG- 
ISTERS AGAINST 
FIXED SURFACE 
SOCKET 



BODY OF BASE SHALL 
PASS THRU A RING OF 
1.098 D 




ALL DIMENSIONS IN INCHES 



1. Scope. The purpose of this standard is to 
establish, for the type of lamp shown, the di- 
mensions essential to interchangeability of 
lamps in projectors. It is not intended to pre- 
scribe either operating characteristics or de- 
tails of design. 

2. Operating Position. Lamps of this type 
are intended to be burned with the axis in an 
essentially vertical position, and with the base 
at the bottom. 

Note 1. These dimensions define the maximum ex- 
cursion of the bulb surfaces from the base axis toward 



the condensing lenses and the mirror at the points 
indicated when the lamp is inserted in a holder which 
rotational!/ positions the lamp as shown in the end 
view of the base. Condensing lenses, the mirror, and 
their mounts must therefore be so located as to insure 
adequate clearance between these parts and the bulb 
surface. 

Note 2. For medium prefocus ring double-contact 
base-up projection lamps, see American Standard 
Dimensions for Projection Lamps Medium Prefocus 
Ring Double-Contact Base-Up Type for 16mm and 
8mm Motion Picture Projectors, PH22.84-1953, or the 
latest revision thereof approved by the American 
Standards Association, Incorporated. 



Approved January 8, 1953, by the American Standards Association, Incorporated 
Sponsor: Society of Motion Picture and Television Engineers 



Univrrut Drrioul CUoi&ralioo 



Copyright 1953 by the American Standards Ass< 
70 East Forty. 6tth Street, New York 17, N. Y. 



elation. Incorporated 



Price, 25 Cent* 



January 1953 Journal of the SMPTE Vol. 60 



71 



American Standard 

Enlargement Ratio for 16mm 
to 35mm Optical Printing 



Keg. V. S. Pal. Of. 

PH22.92-1953 



UDC 778.5=778.13 



In the enlargement printing of 16mm film to 35mm film, a magnifica- 
tion of 2.21 0.01 shall be employed, and the center of the 16mm frame 
as enlarged shall coincide with the center of the 35mm aperture in the 
enlarging printer. 



Note 1. This will mean a scanned area on the 
16mm frame of 0.272 inch : 0.002 X 0.373 inch 
0.002 will be projected through the 35mm projector 
aperture when the print is used in the theater. This 
corresponds to a frame of 0.284 inch X 0.380 inch 
if the 16mm original were projected directly. 

Note 2. The scanned area of the 16mm frame in 
the printer as enlarged to the 35mm camera aper- 
ture is 0.286 inch 0.002 X 0.393 inch 0.002. 
Note 3. Attention of camera users is invited to the 
desirability of using a camera finder matte 0.272 
inch 0.002 X 0.373 inch 0.002 when exposing 
16mm film to be enlarged to 35mm film. 



Note 4. In enlargement from 16mm positive or re- 
versal original to 35mm negative a black frame line 
will result on the final 35mm print. In the case of en- 
largement from 16mm negative directly to 35mm 
print, white frame lines will result. If the height of the 
16mm aperture for enlargement from 16mm negative 
to 35mm print is made 0.300 inch, the resulting aper- 
ture image on the 35mm print will be from 0.660 to 
0.666 inch in height. While the frame line will not be 
entirely black, there would be a black margin on 
either side of the image which would give an addi- 
tional safety factor in projection. 



Approved January 8, 1953, by the American Standards Association, Incorporated 
Sponsor: Society of Motion Picture and Television Engineers 



Universal Decimal Classification 



Price, 25 Cents 



72 



January 1953 Journal of the SMPTE Vol. 60 



73d Semiannual Convention 



Papers are now a first order of business. 
Papers Committee Vice-Chairmen, listed 
in the December Journal, have been active 
with 73d Convention Author's Forms issued 
at the end of December by W. H. Rivers, 
Papers Committee Chairman. Bill Rivers 
has slightly revamped the "Hints to 
Authors," especially to try to get better 
presentations of lantern slides at Conven- 
tions. With the Author's Forms for this 
Convention goes a copy of American Stand- 
ard Dimensions for Lantern Slides, Z38.7- 
1950, to which the American Standards 
Assn. permitted the SMPTE to add some 
recommendations for lettering as a mini- 
mum guidance for readability. 

Ralph Lovell, Program Chairman, al- 
ready has some sessions sketched out, par- 
ticularly for a group of papers about out- 
door theaters. SMPTE's television papers 
for this convention will be scheduled in a 
coordinated plan with the National Associa- 
tion of Radio and Television Broadcasters. 



The NARTB meeting is being held at the 
Biltmore Hotel in Los Angeles on 
April 28-May 1. The SMPTE Conven- 
tion begins on April 27. It is expected that 
many will want to see the NARTB ex- 
hibits. 

The general outline of the 73d Conven- 
tion will be mailed about March 2 as the 
Advance Notice. 

A new feature expected to be introduced 
at this Convention by Convention Vice- 
President J. W. Servies and Local Arrange- 
ments Chairman Vaughn C. Shaner is 
opening of the Registration Desk a day 
earlier, on Sunday, at the Los Angeles 
Statler. 

Now is the time the scheduled deadline 
is February 16 to get Author's Forms in 
for the Los Angeles Program. Forms and 
information are available, always from 
Society headquarters, but better from the 
Papers Committee member in your area or 
organization : 



Chairman: W. H. Rivers, Eastman Kodak Co., 342 Madison Ave., New York 17. 

73>d Convention Program Chairman: Ralph E. Lovell, 2743 Veteran Ave., West Los Angeles 

64, Calif. 

For Washington: J. E. Aiken, 116 N. Galveston St., Arlington 3, Va. 
For New York: Skipwith W. Athey, 201 Spring St., Mt. Kisco, N.Y. 
For Chicago: Geo. W. Colburn, 164 N. Wacker Dr., Chicago 6, 111. 
For 73d Convention High-Speed Photography: Carlos H. Elmer, 41 OB Forrestal St., China 

Lake, Calif. 

For Canada: G. G. Graham, National Film Board of Canada, John St., Ottawa, Canada 
For High-Speed Photography: John H. Waddell, 850 Hudson Ave., Rochester 21, N.Y. 



Papers Committee Members 

James A. Anderson, Alexander Film Co., 

Alexander Film Bldg., Colorado 

Springs, Colo. 
Mark Armistead, 1041 N. Formosa Ave., 

Hollywood 46, Calif. 

D. Max Beard, Naval Ordnance Labora- 
tory, White Oak, Silver Spring, Md. 
Richard Blount, General Electric Co., Nela 

Park, Cleveland, Ohio 
R. P. Burns, Balaban & Katz, Great States 

Theaters, 177 N. State St., Chicago 1, 

111. 
Merle H. Chamberlin, Metro-Goldwyn- 

Mayer Studios, 10202 Washington 

Blvd., Culver City, Calif. 
P. M. Cowett, Dept. of the Navy, Bureau of 

Ships, Washington 25, D.C. 



E. W. D'Arcy, De Vry Corp., 1111 W. 
Armitage Ave., Chicago 14, 111. 

W. H. Deacy, Jr., 231 E. 76 St., New York 
21, N.Y. 

W. P. Dutton, 732 N. Edison St., Arling- 
ton 3, Va. 

Barry T. Eddy, 10569 Selkirk Lane, Los 
Angeles, Calif. 

Karl Freund, 15024 Devonshire St., San 
Fernando, Calif. 

Jack R. Glass, 10858 Wagner St., Culver 
City, Calif. 

R. N. Harmon, Westinghouse Radio Sta- 
tions, Inc., 1625 K St., N.W., Wash- 
ington, D.C. 

Scott Helt, Allen B. Du Mont Laboratories, 
Inc., 2 Main Ave., Passaic, N.J. 



73 



G. E. Heppberger, 231 N. Mill St., Naper- 
ville, 111. 

S. Eric Howse, 2000 West Mountain St., 
Glendale 1, Calif. 

L. Hughes, Hughes Sound Films, 1200 
Grant St., Denver, Colo. 

P. A. Jacobsen, Campus Studios, 100 
Meany Hall, University of Washing- 
ton, Seattle, Wash. 

William Kelley, Motion Picture Research 
Council, 1421 N. Western Ave., Holly- 
wood 27, Calif. 

George Lewin, Signal Corps Photographic 
Center, 25-1135 St., Long Island 
City 1, N.Y. 

Glenn E. Matthews, Research Laboratory, 
Eastman Kodak Co., Rochester 10, 
N.Y. 

Pierre Mertz, Bell Telephone Laboratories, 
Inc., 463 West St., New York 14. 

Harry Milholland, Du Mont TV Network, 
Station WABD, 515 Madison Ave., 
New York 22. 

W. J. Morlock, General Electric Co., Elec- 
tronics Park, Syracuse, N.Y. 



Herbert W. Pangborn, 6512 Orion St., 

Van Nuys, Calif. 
Bernard D. Plakun, General Precision 

Laboratory, Inc., 63 Bedford Rd., 

Pleasantville, N.Y. 
CarlN. Shipman, 9544 Burma Rd., Rivera, 

Calif. 
S. P. Solow, Consolidated Film Industries, 

Inc., 959 Seward St., Hollywood 38, 

Calif. 
J. G. Stott, Du-Art Film Laboratories, 245 

W. 55 St., New York 19. 
W. L. Tesch, Radio Corporation of Amer- 
ica, RCA Victor Div., Front & 

Cooper Sts., Camden, N.J. 
Lloyd Thompson, The Calvin Co., 1105 

Truman Rd., Kansas City 6, Mo. 
M. G. Townsley, Bell & Howell Co., 7100 

McCormick Rd., Chicago 45, 111. 
Allan L. Wolff, Westrex Corp., 6601 

Romaine St., Hollywood 38, Calif. 
Roy L. Wolford, 3434 W. 110th St., 

Inglewood 2, Calif. 



Royal Photographic Society Centenary 



The Royal Photographic Society of Great 
Britain celebrates its Centenary in 1953 
and will hold an International Conference 
on the Science and Applications of Photog- 
raphy in London from Saturday, September 
19, to Friday, September 25, 1953. 

The Conference will cover many aspects 
of the science, technique and applications 
of photography and will be divided into 
sections dealing with: 

I. Photographic Science (including 
theory of latent image and development, 
sensitization, sensitometry, resolving power, 
granularity, properties of photographic ma- 
terials). 

II. Cinematography and Colour Photog- 
raphy. 

III. Technique and Applications of 
Photography (including industrial radiog- 
raphy, photomicrography, spectroscopy, 
aerial photography, photogrammetry, high- 
speed photography, nuclear track record- 
ing, and other physical chemical and bio- 



logical applications; photocopying, appa- 
ratus, process, manipulations). 

IV. Photomechanical Processes. 

V. History, Literature (including ab- 
stracting and documentation) and Train- 
ing in Photography. 

All persons taking an interest in photog- 
raphy or its applications are cordially in- 
vited to attend the Conference, and to sub- 
mit papers for discussion. Titles and indi- 
cations of the scope of such papers should 
be submitted before February 1, 1953. 

Preparation, Presentation and 
Publication of Papers 

It is the intention of the Organizing 
Committee to publish the accepted papers 
soon after the Conference, and to distribute 
preprints as early as possible before its 
opening. Titles and an indication of the 
scope of papers should be submitted before 
February 1, 1953. The full text (in dupli- 
cate) of all addresses, reports and papers 



74 



must be in the hands of the Organizing 
Committee before May 1, 1953. 

In order to allow ample time for discus- 
sion, the authors will be accorded only a 
few minutes to introduce their papers and 
to outline their conclusions. A longer 
speaking time will be granted only for the 
introduction by the Chairman of each sec- 
tion and for specially invited addresses and 
papers. Details on the form of the type- 
scripts and illustrations, on the publication, 
reprinting, etc., will be sent to all intend- 
ing contributors. The Committee has 
absolute discretion as to the acceptance of 
papers. No information of any kind con- 
cerning rejected communications will be 
published. No paper shall be published 
before it has been read at the Conference. 

It is to be emphasized that papers sub- 



mitted for the Conference will receive every 
consideration, but it will not necessarily 
mean that they will be accepted for pre- 
sentation or publication. 

[The requirements of the Royal Photo- 
graphic Society may be of particular inter- 
est to SMPTE authors and committecmen, 
for the sake of comparison.] 

Those who can plan to participate in this 
International Conference should send by 
air mail or cable their advice and abstract 
to Mr. C. S. Brasier, 4 Romney Road, 
Southcourt, Aylesbury, Bucks, England. 
Mr. Brasier is particularly interested in 
high-speed photography but his interest in 
the SMPTE generally should prompt 
whatever cooperation may be feasible on 
the part of any of this Society's members. 



Engineering Activities 



The rules of the American Standards Association require periodic review of all standards 
over three years old for reaffirmation, revision or withdrawal. In accordance with this 
procedure, the Engineering Committees have been participating this past year in an ex- 
tensive review of those cinematographic standards which were issued prior to 1949. The 
status of this activity is presented below. Henry Kogel, Staff Engineer. 



*Std. Title 

.2-1946, 35mm Film-Usage in 

Camera 
.3-1946, 35mm Film-Usage in 

Projector 
.4-1941, Projection Reels for 35mm 

Film 
.5-1947, Dimensions for 16mm 

Double-Perforated Film 



.9-1946, 16mm Double-Perforated 

Film-Usage in Camera 
.10-1947, 16mm Double-Perforated 

Film-Usage in Projector 
.12-1947, Dimensions for 16mm 

Single-Perforated Film 
.15-1946, 16mm Single-Perforated 

Film-Usage in Camera 
.16-1947, 16mm Single-Perforated 

Film-Usage in Projector 
.17-1947, Dimensions for 8mm Film 



Status 

Revision proposed by the Sound Committee, 
approved by the Standards Committee, and is 
now being reviewed by ASA Sectional Committee 
PH22. 

Now being reviewed by the Film Projection Prac- 
tice Committee. 

Revision proposed by 16mm and 8mm Commit- 
tee, approved by Standards Committee, Pub- 
lished in December 1952 Journal for 3-month trial 
and comment. 

Second Draft being prepared within 16mm and 
8mm Committee 

Same as .5 above. 



16mm and 8mm committee now voting on these 
proposed revisions. 

Film Dimensions Committee now voting on pro- 
posed revision. 



* All these standards had the Z22 designation. This is now being replaced by PH 
precede the decimal point in the number of each standard. 



75 



*Std. Title 

.21-1946, 8mm Double-Perforated 

Film-Usage in Camera 
.22-1947, 8mm Double- Perforated 

Film-Usage in Projector 
.23-1941, Projection Reels for 8mm 

Film 
.27-1947, Determining Transmission 

Density of Film 

.28-1946, Projection Rooms and 

Lenses for Theaters 
.31-1946, Definition for Safety Film 

.33-1941, Nomenclature for Electrical 
Filters 

.35-1947, 16-Tooth 35mm Projector 

Sprocket 
.36-1947, Dimensions for 35mm 

Positive Film 

.37-1944, Raw Stock Cores for 35mm 

Film 
.38-1952, Raw Stock Cores for 16mm 

film 
.39-1944, Screen Brightness for 35mm 

Motion Pictures 



.41-1946, Sound Records and Scan- 
ning Area 16mni Sound 
Prints 

.42-1946, 16mm Sound Focusing Test 
Film 

.43-1946, 16mm 3000-Cycle Flutter 
Test Film 

.44.1945, 16mm Multi-Frequency 
Test Film 

.45-1946, 16mm 400-Cycle Signal 
Level Test Film 

.46-1946, 1 6mm Positive Aperture and 
Image Size for Positive 
Prints From 35mm Nega- 
tives 

.47-1946, Negative Aperture Image 

Size 16mm Duplicate Nega- 
tives from 35mm Positive 
Prints 

.48-1946, Printer Aperture Contact 
Printing 16mm Positive 
From 16mm Negatives 



Status 






Revision, purely editorial in nature, proposed 
the 16mm and 8mm Committee, approved by 
the Standards Committee, and is now being re- 
viewed by ASA Sectional Committee PH22. 
Requires revision by 16mm and 8mm Committee. 

Reaffirmed by the Laboratory Practice Commit- 
tee, and now being reviewed by the Standards 
Committee 

Requires revision by Film Projection Practice 
Committee. 

Requires revision. Is being studied by ASA 
Sectional Committee PHI. 

Approval withdrawn by ASA October 1952.^ 
Withdrawal notice published November 1952! 
Journal. 

Now being reviewed by the Film Projection 
Practice Committee. 

Film Dimensions Committee now voting on pro- 
posed revision. Only change is method of indi- 
cating dimension G. 

Reaffirmed by Film Dimensions Committee and 
now being reviewed by Standards Committee. 
Revision was approved October 1952. Published 
in November 1952 Journal. 

Revision proposed by Screen Brightness Com- 
mittee, approved within SMPTE, published for 
trial and comment in May 1952 Journal, and is 
now being reviewed by the Photographic Stand- 
ards Correlating Committee of ASA. 
Revision is being drafted by the Laboratory 
Practice Committee. 

Sound Committee is now voting on proposed 
revision. 

Reaffirmed by the Sound Committee and is now 
being reviewed by the Standards Committee. 

Same as .42 above. 
Same as .27 above. 



Same as .27 above. 



Same as .27 above. 



* All these standards had the Z22 designation. This is now being replaced by PH22 to 
precede the decimal point in the number of each standard. 



76 




*Std. 



Title 



.49-1946, Printer Aperture Contact 

Printing 1 6mm Reversal and 
Color Reversal Duplicate 

.50-1952, Reel Spindles for 16mm 

Projectors 
.51-1946, Intermodulation Tests on 

Variable-Density 16mm 

Prints 

.52-1946, Cross-Modulation Tests on 
Variable- Area 1 6mm 
Prints 

.53-1946, Revolving Power of 16mm 
Projector Reels 



.54-1946, 16mm Travel Ghost Test 
Film 

.55-1947, 35mm Release Prints 



.56-1947, Nomenclature for Film Use 
in Studios and Processing 
Laboratories 



.57-1947, Buzz Track Test Film 16mm 
Reproducers 

.58-1947, Picture Projection Aperture 
35mm Projectors 

.60-1948, Theater Sound Test Film for 
35mm Reproducing Systems 

.62-1948, 35mm Sound Focusing Test 
Film (Lab. Type) 

.65-1948, 35mm Scanning-Beam 
Uniformity Test Film 
(Service Type) 

.66-1948, 35mm Scanning-Beam 
Uniformity Test Film 
(Lab. Type) 

.67-1948, 35mm 1000-Cycle Balancing 
Test Film 

.69-1948, Sound Records and Scan- 
ning Area of Double-Width 
Push-Pull Sound Prints 
(Normal Centerline Type) 

.70-1948, Sound Records and Scan- 
ning Area of Double-Width 
Push-Pull Sound Prints 
(Offset Centerline Type) 



Status 

Is being reviewed by Laboratory Practice Com- 
mittee. 



Reaffirmed by ASA November 1952 and pub- 
lished in the December 1952 Journal. 



Reaffirmed by the Sound Committee and is now 
being reviewed by the Standards Committee. 

Revision proposed by Optics Committee, ap- 
proved within SMPTE, and is now being re- 
viewed by the Photographic Standards Correlat- 
ing Committee. 

Same as .23 above. 

Revision required. Draft is being prepared by 
the Laboratory Practice and Films for Television 
Committees. 

Is being revised by Laboratory Practice Com- 
mittee. 

Standard .57 and the remainder of those listed 
here have been reaffirmed by the Sound Com- 
mittee and are now being reviewed by the 
Standards Committee. 



* All these standards had the Z22 designation. This is now being replaced by PH22 to 
precede the decimal point in the number of each standard. 

77 



Letters to the Editor 



Re: Three-Dimensional Motion Picture Nomenclature 



[from L. Dudley] 

I would like to refer to my letter, and 
Major Bernier's reply thereto, which 
appeared in the Journal for July 1952. 

With regard to a suitable term to define 
those stereoscopic processes which do not 
entail the use of individual viewing devices, 
I think that the term autostereoscopic proc- 
esses is as good as any. This term is in 
fairly general use in England, and equiva- 
lent terms are gaining some ground on the 
continent. Further, on the writer's recom- 
mendation, it has been adopted by the 
British Standards Institute. 

I notice that Major Bernier has repeated 
some of the information, concerning early 
pioneers, which I gave in my own letter, 
but has made one or two errors in this 
connection. For example, the particular 
member of the Ives family who is asso- 
ciated with the year 1902 is, as stated in 
my letter, Frederick Ives (the inventor of 
the parallax stereogram), and not his son, 
Dr. H. E. Ives, as inferred by Major 
Bernier. Dr. H. E. Ives's most important 
contributions to the art lie in his various 
proposals for applying the principle of the 
parallax panoramagram (invented by G. 
W. Kanclt in 1915) to stereo cinematog- 
raphy. 

Referring to the seventh paragraph of 
Major Bernier's letter, it would appear 
that Major Bernier agrees that "accom- 
modation" is the correct term, rather 
than "focus reaction," so it is a little 
difficult to follow his reason for believing 
that the latter term would be more easily 
understood. 

With reference to the comments in para- 
graphs 8 to 10 of Major Bernier's letter, 
here again I am at a loss to follow his 
reasoning. My statement to the effect 
that stereoscopic vision is the net result of 
the various contributing factors is not 
based on a fallacy, for the very good reason 
that my definition of the term is that 
which is generally accepted. Further, the 
fact that binocular vision does not always 
result in the perception of a three-dimen- 
sional (or stereoscopic) image has not been 
in question, so I do not understand why 
Major Bernier cites several examples to 



78 



illustrate this point. By so doing, Major 
Bernier is, in fact, arguing on my side, 
because he is agreeing that, whilst bin- 
ocular vision does not always result in 
the perception of a three-dimensional (or 
stereoscopic) image, we cannot experience 
stereoscopic vision without binocular vision. 
Accordingly, the latter factor must be 
regarded as one of the contributory causes 
of the former. 

I am familiar with the work of Hardy and 
Perrin, to which Major Bernier refers, 
and agree that in most circumstances the 
faculties of accommodation and con- 
vergence are interdependent. Such inter- 
dependence exists during the viewing of 
motion pictures, as can be demonstrated 
experimentally by photographic methods. 
In my previous letter, when discussing the 
phenomena causing cinema patrons to 
make periodic, momentary efforts to 
accommodate for the "apparent" plane of 
the image, I stated that this "is sometimes 
the cause of headaches amongst elderly 
cinema patrons, whose ocular sensory 
organs and muscles are, naturally, less 
responsive than those of younger people." 
I am very surprised, therefore, to note 
Major Bernier's comment that he cannot 
agree with this "since it is common 
knowledge in ophthalmic practice that 
they lose their power of accommodation 
as a result of progressive hardening of the 
crystalline lens as they grow older." Here, 
again, it would appear that Major Bernier 
is not really disagreeing with me, because 
we are, of course, both in agreement about 
the progressive deterioration, with age, 
of the power of accommodation. The 
important point is that the effort to ac- 
commodate does not undergo the same, 
progressive deterioration, and it is the 
persistence of this effort, regardless of the 
fact ,that the organs concerned are no 1 
longer fully responsive, which results in 
strain. 

I note that Major Bernier claims to have 
overcome completely the time parallax 
problem associated with the alternate frame 
principle. I would therefore draw atten- 
tion to the fact that, by definition, an 
alternate frame system is one in which the 



"left-eye" and "right-eye" views are 
recorded alternately. Accordingly, the 
phenomenon of time parallax is inherent 
in such systems. In order to eliminate 
time parallax, the two components of each 
successive stereoscopic pair must be re- 
corded simultaneously, which means, of 
course, abandonment of the alternate 
frame principle. 

Referring to the penultimate paragraph 
of Major Bernier's letter, my statement 
that "the minimum rate of occultation 
necessary to prevent the occurrence of 
objectionable flicker is about 24 per second" 
is quite correct. In my previous letter I 
was not referring to the rate of occultation 
which would be acceptable in a practical 
system, as the acceptable rate is influenced 
by screen brightness, picture contrast and 
other factors. In a practical system the 
rate of occultation must be sufficiently 
high to cope with the most severe condi- 
tions, necessitating doubling or trebling 
the minimum rate. 

September 16, 1952 L. P. C. J. Dudley 
Stereoptics Limited 
The Laboratory 
Odeon Theatre 
Kensington High St. 
London, W. 8. 



[from John A. Norling] 

I have reviewed with interest Mr. L. 
Dudley's letter of 30 August 1951, Major 
Robert V. Bernier's letter in reply thereto 
and Mr. Dudley's comments of 16 Sept. 
1952 on Major Bernier's later communica- 
tion. 

These letters confirm my opinion that 
the Stereoscopic Art needs an authoritative 
nomenclature, a nomenclature that will 
make it possible for all authors who will 
follow it to "speak a common language." 
It is my hope that the Stereoscopic Com- 
mittee will soon produce a Glossary of 
Terms acceptable to all workers in the 
field. There exist many phases of the 
art which have been given various terms 
by different people and there are no 
textbooks on ophthalmology and optometry 
which completely cover the field. How- 
ever, we should not change a term from 
its established use merely because we think 
we can express ourselves more clearly than 
the textbooks have succeeded in doing. 



I agree with Mr. Dudley that "accom- 
modation" should be used rather than 
Maj. Bernier's "focus reaction." The terra 
"autostereoscopic process (es)" is much to be 
preferred to the loose term "composite 
process (es)." Composite processes of 
photography are not part of the stereo- 
scopic art; in fact the familiar "Composi- 
graph," employed extensively in the past 
as a journalistic stunt, is something that 
would result only in a confusing stereo 
movie, if it could be made at all. 

An interesting thing about projected 
stereo is that it calls for a relaxation of 
accommodation but an active use of the 
muscles employed in convergence. For 
this reason it seems advisable to use 
extreme restraint in employing photo- 
graphic stunts that require the viewer of 
projected stereo excessive use of his faculty 
of convergence. Eyestrain may arise if 
wide uncoupling of accommodation and 
convergence are demanded of people. 

I agree with Mr. Dudley that the time 
parallax problem associated with the 
alternate-frame principle is a serious one, 
particularly if the individual members of 
the stereo pairs are photographed alter- 
nately. But even if the individual members 
are photographed simultaneously, alternate 
projection will almost certainly result in 
eyestrain, regardless of the projection 
frequency. 

The eyes, or perhaps more properly the 
visual centers, do not like the delivery of 
a picture to one eye during an occultation 
of the other eye. I have observed symp- 
toms of nausea even at projection of 48 
frames a second, 24 to each eye, and a 
flicker frequency of 192. I have used the 
term "differential flicker," for want of a 
better term, in discussing the particular 
problems in alternate-frame projection. 
Flicker fusion frequency (fff) is a very 
important matter in ophthalmology and 
medical practice. Detection of the change 
in fff in an individual is often of great 
value in diagnosis. 

In connection with fff, research has 
demonstrated that the average is around 
45 to 48 cycles/sec and doesn't vary much 
with age, but tests for fff are made with a 
beam of light subtending one degree or 
less and covering only the fovea, where, 
as we know, there is less sensitivity to 
rapid changes in light than in the outer 
regions of the retina. The motion picture 



79 



is seen by quite a large area of the retina 
and a fairly bright picture usually has a 
detectable flicker at 96 interruptions a 
second, 48 periods of brightness and 48 
periods of darkness, as occur in projection 
with a two-bladed shutter at 24 frames /sec. 
A very limited study of the flicker problem 
has convinced me that alternate-frame 
projection of stereo has serious drawbacks. 
Therefore I cannot agree with Major 
Bernier's conclusions nor with Mr. Dudley's 
statement (in his 30 Aug. 1951 letter) that 



"It is readily demonstrable that the mini- 
mum rate of occultation to prevent the 
occurrence of objectionable flicker is about 
24 frames /sec, etc.," even though he 
confines this to planoscopic projection and 
even though he qualifies this statement in 
the last paragraph of his letter of Sep- 
tember 16, 1952. 



October 1, 1952 



John A. Norling 
245 W. 55 St. 
New York 19, N.Y. 



New Members 



The following members have been added to the Society's rolls since those last published. 
ignations of grades are the same as those used in the 1952 MEMBERSHIP DIRECTORY. 



The des- 



Honorary (H) 



Fellow (F) 



Active (M) 



Associate (A) 



Student (S) 



Agar, H. V., Sales, Watland Brothers. Mail: 

535 North Brainard, La Grange Park, 111. (A) 
Arthur, Hal, University of Southern California. 

Mail: 8569 Nash Dr., Los Angeles 46. (S) 
Barkes, Gordon S., Projectionist, WBKB Tele- 
vision Station. Mail: 4170 West Nelson, 

Chicago 41, 111. (A) 
Belsky, Clarence J., University of Southern 

California. Mail: 430^ South Burlington 

Ave., Los Angeles 5, Calif. (S) 
Benns, William E., Jr., Consulting Radio and 

Television Engineer, 3738 Kanawha St., 

N.W., Washington, D.C. (M) 
Bonner, Bob, University of Southern California. 

Mail: 2326 Scarff St., Los Angeles 7. (S) 
Brokaw, Edgar L., Jr., Lecturer (Motion 

Picture Editing), University of California at 

Los Angeles. Mail: 129 South Oakhurst 

Dr., Beverly Hills, Calif. (A) 
Brokop, Robert J., University of Southern 

California. Mail: 1130 West 36 St., Los 

Angeles 7, Calif. (S) 
Browne, Robert A., University of Southern 

California. Mail: 836 W. 41 St., Los 

Angeles 37, Calif. (S) 
Caldwell, S. W., President, S. W. Caldwell, 

Ltd., 150 Simcoe St., Toronto, Ontario, 

Canada. (A) 
Caras, Roger A., University of Southern 

California. Matt: 930 W. 36 St., Los 

Angeles 7, Calif. (S) 
Cleveland, George, Production Manager, 

Great Commission Films. Mail: 526 Moreno 

Ave., Los Angeles 49, Calif. (A) 
Colman, Joel E., University of California at 

Los Angeles. Mail: 3267 Sepulveda Blvd., 

Apt. 1, Los Angeles 34, Calif. (S) 
Cowles, William E., Mechanical Engineer; 

Groupleader, Engineering Visual Aids., 

General Electric Co. Mail: 1556 Clifton 

Park Rd., Schenectady, N.Y. (A) 



Croy, Harlan P., General Manager and Treas- 
urer, Film Arts Corp. Mail: 1032 N. 

Sixth St., Milwaukee, Wis. (M) 
Cunningham, Clairdon E., Research Psycholo- 
gist (Experimental), U.S. Navy Electronics 

Laboratory. MaU: 3628 Charles St., San 

Diego 6, Calif. (M) 
Day, I. M., Supervisor, Northern Electric Co., 

Ltd., P.O. Box 6124, Montreal, Quebec, 

Canada. (M) 
Desrosiers, Robert, Assistant Film Editor, 

Canadian Broadcasting Corp. (Television). 

Mail: 3757 Carlton Ave., Montreal 26, 

Quebec, Canada. (A) 
Dill, James M., Electronic Engineer, UM&F 

Manufacturing Corp. Mail: 12215 Victoria, 

Los Angeles 34, Calif. (A) 
Dillard, Albert E., University of Southern 

California. Mail: 1820 West 38 St., Los 

Angeles 62, Calif. (S) 
Ebron, Bonifacio M., Jr., University of Southern 

California. Mail: 837 West 36 Place, Los 

Angeles, Calif. (S) 

Elias, T. J., Control Chemist (Solutions), Tech- 
nicolor M.P.S. Mail: 4943 Densmore, 

Encino, Calif. (A) 
Filizola, Vincent F., Television Engineer, 

Paramount Television Productions, Inc. 

(KTLA). Mail: 5327 Loma Linda Ave., 

Hollywood 27, Calif. (M) 
Friesen, Dietrich P., University of Southern 

California. Matt: 942 West 34 St., Los 

Angeles 7, Calif. (S) 
Galminas, Dominic, Director, Cameraman, 

Editor, Applied Physics Laboratory, Johns 

Hopkins University. Mail: 9805 Warren 

St., Silver Spring, Md. (A) 
Ganon, Bob R., Production Manager, TV 

Ads, Inc., 3839 Wilshire Blvd., Los Angeles 5, 

Calif. (M) 



80 



Garcia, Gilberto E., University of Southern 
California. Mail: 1201 South Third Ave., 
Los Angeles 19, Calif. (S) 

Ghosh, Ishan, Recordist, c/o Kardar Pro- 
ductions, Parel, Bombay, India. (A) 

Gill, George H., Television Lighting Sales 
Engineer, Kliegl Bros. Mail: 13 Smith St., 
Glen Head, N.Y. (A) 

Green, F. A., Technical Officer, Audio-Visual 
Aids, International Civil Aviation Organiza- 
tion, 716 International Aviation Bldg., 
Montreal, Canada. (M) 

Griffin, William C., Photographic Technologist, 
U.S. Naval Ordnance Test Station. Mail: 
P.O. Box 637, China Lake, Calif. (M) 

Gunzburg, M. L., President, Natural Vision 
Corp. Mail: 1710 North La Brea Ave., 
Hollywood 46, Calif. (A) 

Gyiaung, Tin, University of Southern Cali- 
fornia. Mail: 837 36 Place, West Aeneas 
Hall, Los Angeles 7, Calif. (S) 

Harvey, Douglas G., University of Southern 
California. Mail: 1846 South Cochran 
Place, Los Angeles 19, Calif. (S) 

Head, Howard T., Consulting Radio Engineer, 
A. D. Ring & Co., 342 Munsey Bldg., Wash- 
ington, D.C. (A) 

Hedburn, Paul W., Motion Picture Laboratory 
Manager, Atlas Film Corp. Mail: 646 
Adams St., Oak Park, 111. (A) 

Hedwig, Gordon W., Telecast Films, Inc. 
Mail: 2 Keenan Place, Garden City, N.Y. 
(M) 

Holdeman, Don, Cine-Technician, Public 
Relations Dept., Arabian American Oil Co., 
Dhahran, Saudi Arabia. (A) 

Holleran, J. Vinson, Vice-President, McGeary- 
Smith Laboratories, Inc., 1905 Fairview Ave., 
N.E., Washington, D.C. (A) 

Ives, George M., Television Engineer, Main- 
tenance Supervisor, American Broadcasting 
Co. Mail: 4221 Arthur Ave., Brookfield, 
111. (M) 

Jekste, Alberts Z., Managing Director, Atlantic 
Films & Electronics, Ltd., 22 Prescott St., 
St. John's, Newfoundland. (M) 

Klaeger, Robert H., Vice-President in charge of 
Television Production, Transfilm, Inc., 35 
W. 45 St., New York 19, N.Y. (M) 

Lapenieks, Vilis, University of Southern 
California. Mail: 2905 South Hoover St., 
Los Angeles, Calif. (S) 

Leavenworth, William, University of Southern 
California. Mail: 907 West 28 St., Los 
Angeles, Calif. (S) 

LeGault, Joseph W., University of Southern 
California. Mail: 612 West 115 St., Los 
Angeles 44, Calif. (S) 
Lewis, Vernon, Motion Picture Producer, 71 

W. 45 St., New York 19, N.Y. (M) 
Lindholm, George W., Jr., Photo Unit Chief, 
Argonne National Laboratory. Mail: 1742 
E. 83 PI., Chicago 17, 111. (A) 
Loughren, Arthur V., Engineer, Director of 



Research, Hazeltine Corp. Mail: 22 Broad- 
lawn Ave., Great Neck, N.Y. (M) 
Marcus, Omar, Technical Consultant, Tri-Art 
Color Corp., 245 W. 55 St., New York 19, 
N.Y. (M) 

McBrien, Donald G., Research Associate in 
Photography, Optical Research Laboratory, 
Boston University. Mail: 28 Orchard Rd., 
Swampscott, Mass. (A) 

Price, Robert S., General Engineer, Naval 
Ordnance Laboratory. MaU: 12 Mason 
Rd., Indian Head, Md. (A) 
Richards, A. H., University of Southern Cali- 
fornia. Mail: 5356 Lexington, Apt. 108, 
Hollywood 29, Calif. (S) 
Richardson, Norman, Photographer, Sandia 

Corp. Mail: Box 529, Brawley, Calif. (A) 
Robertson, Robert B., Laboratory Technician, 
Consolidated Film Industries. Mail: 1622 
North Dillon St., Los Angeles 26, Calif. (M) 
Rouden, Manzia V., Motion Picture Technical 
Advisor, U.S. Air Force. Mail: Star Route, 
Santa Rosa, Fla. (A) 

Sherman, Mendel, University of Southern 
California. Mail: 2625^ Ellendale Place 
Los Angeles 7, Calif. (S) 

Snook, Mary Jean, Research Librarian, Techni- 
color Motion Picture Corp., 6311 Romaine 
St., Hollywood 38, Calif. (A) 
Stein, Morton, Sales, Ray Mercer & Co. Mail: 
9075 W. Pico Blvd., Los Angeles, Calif. (A) 
Weitz, Loyal, University of Southern California. 
Mail: 1909 Farrell, Redondo Beach, Calif. 
(S) 

Whipple, Paul E., University of Southern Cali- 
fornia. Mail: 3436 North Earle Ave., 
Rosemead, Calif. (S) 

Wihtol, Constantine A., New York University. 
Mail: 108-10 66 Ave., Forest Hills, N.Y. 
(S) 

Williams, Marshall A., Electronic Engineer, 
Philco Corp., 260 South Beverly Dr., Beverly 
Hills, Calif. (M) 

Williamson, Harold G., Instrumentation Engi- 
neer, Vitro Corporation of America. Mail: 
General Delivery, Fort Walton, Fla. (A) 
Wilson, James V., Chief Engineer, Film Labora- 
tories of Canada, Ltd. Mail: 289 Forman 
Ave., Toronto, Ontario, Canada. (M) 
Wolff, Leonard E., Audio Engineer, The Hous- 
ton Post Co., KPRC-TV. Mail: 11401 
O'Donnell Dr., Houston 22, Tex. (A) 

CHANGES IN GRADE 

Florman, Arthur, (A) to (M) 
Goldberg, Morris M., (A) to (M) 
Kuzmanov, Alexander B., (S) to (M) 
Ricci, Eduardo J., (S) to (A) 

DECEASED 

Parshley, Charles W., Chief Projectionist, 
University Theater, Harvard Sq., Cambridge, 
Mass. (A) 



81 



Membership Service Questionnaire 



Following through after the extensive dis- 
cussion about cost, type and quality of 
membership service at the October 1952 
Board of Governors Meeting, Society 
headquarters mailed a questionnaire early 
this month, with the annual membership 
dues invoices, to all members, except Stu- 
dents, in the United States. The earlier 
mailing date for invoices to members out- 
side the United States did not permit send- 
ing questionnaires to them. 

Within the ensuing two weeks about six 
hundred members replied. The volume of 
returns is gratifying and confirms the fact 



that the SMPTE is a membership society. 
The exactness and thoughtfulness of the 
replies will add up to a valuable guide to 
confirm some, and modify other, SMPTE 
policies. 

If you have not returned your question- 
naire, please send it along. The tally for a 
report to the Board of Governors and in the 
Journal to the membership will not be 
closed for a few weeks. Comments and 
suggestions are always gratefully received, 
especially suggestions and leads for techni- 
cal papers or new products items. V.A. 



Book Review 



Musical Engineering 

By Harry F. Olson. Published (1952) by 
McGraw-Hill, 330 W. 42 St., New York 36. 
i-ix + 357 pp. + 11 -pp. index. 303 illus. 
6 X 9 in. Price $6.50. 

An encyclopedic volume, even a com- 
pressed one, cannot help but strike a re- 
sponsive chord within a wide range of 
engineers when it is as carefully prepared 
as this one. The engineering bases for 
approaching music will be nothing new to 
the acoustics engineer but others will find 
the engineering principles and the sur- 
rounding meat of the text often of great 
interest and nicely, if not too simply and 
disarmingly, presented. The attention of 
the general reader and the acoustics engi- 
neer alike may be held by the wealth and 
width of this survey. The acoustics 
engineer, and many of his fellow engineers 
in other fields, will find ample biblio- 
graphical references throughout to guide 



them to detailed studies. As far as 
periodicals go, the Journal of the Acoustical 
Society is most often cited, with this Journal 
a strong second in parts of the book. 

For fuller technical explanations, the 
author often gives as a general reference 
his earlier book, Elements of Acoustical 
Engineering. The author is director of the 
Acoustical Laboratory, RCA Laboratories, 
Princeton, N.J. 

To carry out its aim an engineering 
treatment of the interrelated subjects of 
music, musical instruments, acoustics, 
sound reproduction and hearing the 
book is organized in these nine chapters: 
Sound Waves; Musical Terminology; 
Scales ; Resonators and Radiators ; Musical 
Instruments; Characteristics of Musical 
Instruments; Properties of Music; Thea- 
ter, Studio and Room Acoustics; and 
Sound Reproducing Systems. The numer- 
ous illustrations and the index contribute 
to the book's worthiness. V.A. 



Meetings 



American Institute of Electrical Engineers (Symposium on the Science of Music and Its 

Reproduction 4th Lecture), Feb. 20, Engineering Societies Bldg., New York, N. Y. 

National Electrical Manufacturers Association, Mar. 9-12, Edgewater Beach Hotel, 

Chicago, 111. 

Society of Motion Picture and Television Engineers, Southwest Subsection Meeting, 

Mar. 16, Fort Worth, Tex. 



82 



Inter-Society Color Council, Annual Meeting, Mar. 18, Hotel Statler, New York, N. Y. 
Optical Society of America, Mar. 19-21, Hotel Statler, New York, N.Y. 

American Physical Society, Joint Meeting with APS Southeastern Section, Mar. 26-28, 

Duke University, Durham, N.C. 

Symposium on Modern Network Synthesis, planned by Polytechnic Institute of Brooklyn, 
Apr. 16-18, Auditorium of Engineering Societies Bldg., New York 

International Symposium on Nonlinear Circuit Analysis, Apr. 23-24, information from 
Microwave Research Inst., 55 Johnson St., Brooklyn 1, N.Y. 

73d Semiannual Convention of the SMPTE, Apr. 27-May 1, Hotel Statler, Los Angeles 

National Association of Radio and Television Broadcasters, 7th Annual Conf., Apr. 28- 

May 1, Ambassador Hotel, Los Angeles 

American Physical Society, Apr. 30-May 2, Washington, D.C. 

Acoustical Society of America, May 7-9, Hotel Warwick, Philadelphia, Pa. 

Society of Motion Picture and Television Engineers, Southwest Subsection, May 20, 

Dallas, Tex. 

American Physical Society, June 18-20, Rochester, N.Y. 

American Institute of Electrical Engineers, Summer General Meeting, June 29- July 3, 

Atlantic City, N.J. 

Biological Photographic Association, 23d Annual Meeting, Aug. 31-Sept. 3, Hotel Statler, 

Los Angeles, Calif. 

The Royal Photographic Society's Centenary, International Conference on the Science 
and Applications of Photography, Sept. 19-25, London, England 

74th Semiannual Convention of the SMPTE, Oct. 4-9, Hotel Statler, New York 

Audio Engineering Society, Fifth Annual Convention, Oct. 14-17, Hotel New Yorker, 

New York, N.Y. 

Theatre Equipment and Supply Manufacturers' Association Convention (in conjunction 
with Theatre Equipment Dealers' Association and Theatre Owners of America), 

Oct. 31-Nov. 4, Conrad Hilton Hotel, Chicago, III. 

Theatre Owners of America, Annual Convention and Trade Show, Nov. 1-5, Chicago, 111. 

National Electrical Manufacturers Association, Nov. 9-12, Haddon Hall Hotel, Atlantic 

City, N.J. 

75th Semiannual Convention of the SMPTE, May 3-7, 1954 (next year), Hotel Statler, 

Washington, D.C. 

76th Semiannual Convention of the SMPTE, Oct. 18-20, 1954 (next year), Ambassador 

Hotel, Los Angeles 



Employment Service 



Position Wanted Position Available 

Resigning Feb. 1 as gen. mgr., charge Permanent position in Southwest for 

of production, large southern film studio. experience d motion picture cameraman; 

5 yrs. experience as prod, mgr, editor mugt ^ ^ interior and exterior 

and cameraman, 16mm and 35mm. .... ,. T .. , ... 

Married, 37, college grad. References foota S e *> m ? icate ****?' nto 

and resume on request Harlan H. giving resume of professional experience, 

Mendenhall, 1609 Blodgett, Houston 4, to Susong Agency, 524 Commercial Bldg. 

Tex. Dallas, Tex. All replies confidential. 

83 



New Products 



Further information about these items can be obtained direct from the addresses given 
As in the case of technical papers, the Society is not responsible for manufacturers' state- 
ments, and publication of these items does not constitute endorsement of the products. 

A collapsible three-wheel camera dolly 

has been designed to fold into a case 20 X 
20 X 36 in. It is made of cast aluminum. 
The center mount casting provides a hook 
for optional use of the tie-down chains 
when using standard or baby tripods, and 
additional baby tripod point holders are 
provided. Extra-wide rubber wheels have 
been used to prevent side sway. The new 
dolly has floor hand jackscrews for leveling 
or stationary position, foot tread plates for 
the cameraman and the assistant camera- 
man, adjustable seat for the operator, a 
removable steering handle and a lock for 
in-line steering. The dolly is manufactured 
by National Cine Equipment, Inc., 209 W. 
48 St., New York 36, N.Y. 




A new motion picture projection lamp 

has been developed by the Westinghouse 
Lamp Division, with the assistance of Bell 
& Howell engineers. Reported to improve 
home motion picture screen light by as 
much as 20%, it is to be incorporated in 
Bell & Howell's projectors. Increased 
efficiency is reported achieved by a more 
compact biplane filament made by tighter 
winding and closer spacing of the coils 
which in turn is made possible by Westing- 
house's patented Floating Bridge, a sup- 
porting and guiding device for the coils. 
An improved quality of filament wire is 
also cited as a factor in increasing the out- 
put and life of the lamps which are now 
made in 500- and 750-w sizes. A 1000-w 
lamp is being studied as a possible future 
development. 



The intensity of light from mercury-arc 
lamps can now be stabilized by photocell 
control in a combination recently developed 
by Hanovia Chemical & Mfg. Co., 100 
Chestnut St., Newark, N.J. This new light 
source is marketed for use in photochemical 
research and motion picture printing. A 
photoelectric cell in the power supply 
through an electronic circuit, controls the arc 
current in a range from maximum to about 
10% of maximum. These wide limits are 
attainable because the heat of the electric 
arc is no longer utilized to maintain the 
internal vapor pressure, which now de- 
pends on auxiliary heating elements. 



SMPTE Officers and Committees: The roster of Society Officers and the 
Committee Chairmen and Members were published in the April 1952 Journal. 



84 



Rapid Drying of Normally Processed 
Black-and-White Motion Picture Film 

By F. DANA MILLER 



The introduction of rapid drying technique for rapidly processed film suggests 
its possible application for the drying of normally processed motion picture 
films. Consideration of the drying process seems to indicate that hot impinged 
air should be the most satisfactory method of commercial practice. Experi- 
ments were made on a laboratory machine in which the film passes between 
two parallel air chambers. Small holes in the sides of the chambers facing 
the film permit hot air to impinge on both sides of the film. Air temperatures 
ranging from 125 F to 210 F and air velocities of 2,000 to 6,000 fpm were used. 
Eastman Fine Grain Release Positive Safety Film was dried in 10 sec on labora- 
tory equipment of this type and Eastman Plus X Panchromatic Negative Safety 
Film was dried in 16 sec. The physical properties of several films dried in 
this equipment were quite similar to the properties of conventionally dried 
films, but this is not true for all types of film. Rapid driers can be extremely 
compact and their power requirements should be no greater than for con- 
ventional driers. 



I 



N REGENT YEARS several factors have 
combined to focus new attention on the 
problem of drying motion picture films. 
The gradual increase in processing 
machine speeds without change in the 
processing or drying times has led to 
longer and longer thread-ups in the wet 
and dry ends of the machines. It was 
the impetus of the television industry 
which led to the first radical change in 
the processing and drying of film. To 
satisfy television's requirements a tech- 
nique for processing and drying film 



Presented on October 8, 1952, at the So- 
ciety's Convention at Washington, D.C., 
by F. Dana Miller, Manufacturing Experi- 
ments Div., Eastman Kodak Co., Rochester 
4, N.Y. 



at the rate of 90 fpm in 30 sec was 
perfected. Of this cycle only 5 sec are 
used for drying. It must be remembered 
that special high-temperature processing 
is used and its application is limited 
to a few types of films. However, 
barring undesirable effects on the film, 
the drying method might be used for 
any or all types of black-and-white film 
processed in the normal manner. 

Motion picture film defects associated 
with insufficient or excessive drying are 
tackiness, high positive, curl, buckle, 
thermal "in-and-out" of focus, spokiness, 
and flute. These defects and their 
causes were described by Carver, Talbot 
and Loomis. 1 In the conventional type 
of driers, it is not particularly difficult 



February 1953 Journal of the SMPTE Vol. 60 



85 



to avoid the drying conditions which 
produce these defects. While it is 
desirable to dry films to the point where 
the moisture remaining in them is the 
same as the moisture which they would 
have when in equilibrium with air at 
70 F and a relative humidity (R.H.) in 
the range from 45% to 60%, a survey of 
actual laboratory practice has shown 
that the difficulties noted previously are 
not encountered when the equilibrium 
moisture content ranges from as low as 
30% R.H. to as high as 60% R.H. The 
usual type of film drier uses air in this 
relative humidity range at temperatures 
below 100 F. This automatically pre- 
vents excessive over-drying. Under- 
drying is avoided by providing a liberal 
factor of safety in the drying time. The 
very nature of the rapid-drying processes 
precludes both of these safeguards to 
maintaining the drying within the 
proper range. 

It was felt desirable not only to de- 
termine if black-and-white film, normally 
processed, could be dried rapidly without 
harmful effects but in what range the 
equilibrium moisture content of the dried 
film must fall to insure satisfactory 
results. Realizing that the very nature 
of a rapid-drying process would preclude 
the automatic safeguard of over-drying 
which is present in conventional ma- 
chines and that a liberal factor of safety 
on the drying time would partially 
defeat the purpose of rapid drying, it 
was decided that performance data 
should be obtained on a type of drier 
considered the most suitable for opera- 
tion in conjunction with conventional 
processing machines in commercial 
laboratories. This work, therefore, in- 
cludes a brief discussion of various 
methods of obtaining rapid drying, 
several of which were investigated 
by Ives and Kunz, 2 as well as data on 
drying film under a variety of air 
conditions. 

The term "rapid drying" is used, of 
course, in relation to the usual drying 
times obtained in commercial practice 



on conventional machines which for 
positive films range from 8 to 20 min 
and from 15 to 40 min for negative 
films. 8 However, any method which 
resulted in shortening this time to any 
extent might be called "rapid." For 
the purposes of this discussion, however, 
we propose to limit it as applying only 
to those methods which reduce the dry- 
ing times to 10% or less of that obtained 
in the best commercial practice. For 
positive films this means a maximum 
drying time of about 1 min and a 
maximum of 1^ min for negative films. 

Elementary Drying Principles 

Before discussing the various methods 
of drying, a few of the fundamental 
principles of drying ought to be empha- 
sized. The removal of water from a 
material can be accomplished in two 
ways. If the water is standing as fine 
drops on a surface, it can be mechani- 
cally removed by such means as scraping, 
shaking, centrifuges, or air knives. The 
other method is by evaporation. If the 
material containing the water is hygro- 
scopic, then only the surface water 
can be removed mechanically and the 
remainder must be removed by evapora- 
tion. If the material to be dried 
contains more water than it would if 
it were in equilibrium with the air 
around it, water will be slowly evapo- 
rated from it until equilibrium condi- 
tions are reached. This natural evapo- 
ration is a result of the water-vapor 
pressure in the material being higher 
than the vapor pressure of the water in 
the air. The change is slow because 
only temperature convection currents 
cause air movement near the material 
and therefore the water must diffuse 
through the air. Diagrammatically it 
looks as shown in Fig. 1, in which the 
layer of air above the material is shaded 
to show concentrations of water vapor. 
The air at the surface is saturated and 
the surface of the material is at the wet- 
bulb temperature of the air. This 
saturation drops off with the distance 



February 1953 Journal of the SMPTE Vol. 60 



away from the material. If the volume 
of air surrounding the material is large 
with respect to the material, the water 
absorbed by the air will not be an 
appreciable quantity and the relative 
humidity change will be negligible. If 
the air quantity is small, then the ma- 
terial and the air will come to equilib- 
rium at a higher relative humidity. 
The transfer requires a certain amount 
of heat to vaporize the water. With a 
limited amount of air this heat loss can 
be detected in the cooling of the air. 

Here, then, is elementary evaporation. 
It is frequently referred to as constant- 
rate drying. It is typified by evapora- 
tion from a free water surface and is 
practically independent of the material 
being dried. Two ways by which it can 
be accelerated are as follows: one is 
artificially to increase the air circulation 
over the material surface in order to 
decrease the thickness of the air layer 
through which the water must diffuse. 
The air can be blown perpendicular to 
the surface or parallel to it. Either 
method will reduce the depth of the air 
film and therefore increase the rate of 
evaporation. The higher the air velocity 
used the greater the rate of evaporation. 
Figure 2 is a diagram of air flow parallel 
to a sheet of film which illustrates how 
air velocity helps the diffusion of the 
moisture in the air and therefore in- 
creases the rate of evaporation. In the 
range of velocities with which this 
paper is concerned the air flow is pre- 
dominantly turbulent, that is, its average 
velocity is high enough so that it moves 
along with swirls and eddies rather 
than in a streamline manner. 

As the distance from a point in the air 
stream near the film decreases, the drag 
of the film surface slows the turbulent 
air flow, forcing it first into streamline 
flow and finally to zero. The air velocity 
is of no assistance in diffusing the water 
vapor through this quiescent layer of 
air including that which is moving quite 
slowly, it has merely reduced the depth 
of the stagnant layer. The static diffu- 




Fig. 1. Evaporation from a free-water 
surface in still air. The degree of satura- 
tion is indicated by the concentration of 
the dots. 




Fig. 2. Evaporation from a free-water 
surface in turbulent air flowing parallel 
with surface. The flow lines indicate 
the mixing and dispersion of the mois- 
ture-laden molecules in the turbulent 
Zone A. The stagnant layer is Zone B. 

sion forces must force the vapor through 
the layer but the rate at which it can 
be done is a function of the depth of the 
layer. Raising the air velocity will 
therefore result in higher rates of evapo- 
ration. 

Perpendicular flow of air on the sur- 
face, usually called impinged air, pro- 
duces similar effects. Indications are 
that up to 6500 fpm impinged air will 
be most effective in increasing rates of 
evaporation, but above this point parallel 
air flow will be more effective. 

The second method of increasing the 
rate of evaporation is to increase the 
difference in vapor pressure between the 
water on the surface of the film and the 
moisture in the air. Providing the film 
receives no heat by radiation or con- 



F. Dana Miller: Rapid Film Drying 



87 



duction, the water on the surface of the 
film will be approximately at the wet- 
bulb temperature and its vapor pressure 
will therefore be the same as for saturated 
air at this wet-bulb temperature, while 
the air itself will have a vapor pressure 
corresponding to the dew point tempera- 
ture. 4 This difference in vapor pressure 
is sometimes referred to as the "driving 
force" and is roughly proportional to 
the difference between wet- and dry- 
bulb temperatures. When the drying 
air is heated the vapor-pressure difference 
is increased causing a higher rate of 
evaporation. A higher surface tempera- 
ture with the accompanying increase 
in rate of evaporation can also be ob- 
tained by heating the material. 

The foregoing remarks have been 
made with respect to the constant-rate 
phase of the drying of a material. In 
the drying of many materials, and film 
is one of them, the final drying is not 
at a constant rate but at a falling rate. 
This is a result of the rate of moisture 
diffusion through the film to the surface 
being slower than the potential rate of 
evaporation. The same factors, air 
velocity and vapor-pressure difference, 
affect the drying in this stage but their 
action is modified by the diffusion and 
equilibrium moisture characteristics of 
the film. In the rapid drying of film as 
much as two-thirds of the total drying 
time may be in the falling-rate phase. 
As a result the type and thickness of 
both support and emulsion have a large 
effect on the drying time of the film 
under a given set of drying conditions. 

As a brief summary, then, it can be 
said that drying is a heat transfer as 
well as an evaporation or mass transfer 
operation. In the drying of motion 
picture film it is partially done at a 
constant rate of moisture removal and 
partially at a falling rate. In both 
phases the vapor-pressure difference and 
air velocity will affect the speed of 
drying, but in the latter phase the 
characteristics of the film will also 
affect this speed. 



Special Requirements for a 
Commercial Rapid Film Drier 

In considering possible methods of 
rapid drying which might be used in the 
design of a commercial film drier there 
are a few special limitations and re- 
quirements which do not arise in the 
design of conventional machines which 
should be considered. Recognizing the 
fact that short drying times are fre- 
quently obtained at elevated tempera- 
tures, it was realized that it would be 
extremely difficult to control a film drier 
so that the film did not approach the 
high temperatures to which it was 
subjected. Because the film support 
softens at 240 F, it is considered im- 
portant that the film drier be so de- 
signed that the film is never subjected 
to this temperature while in the dry 
state. Some films are made quite 
brittle if subjected to temperatures 
much lower than this and the tempera- 
ture limit on driers for these films 
should be set accordingly. 

In the case of film breaks or other 
accidental stoppages the heat source in 
the machine must not constitute a fire 
hazard. 

The machine should have some ready 
means of adjusting the drying rate, 
independently of the film speed. 

A satisfactory rapid commercial film 
drier must be economical and reliable. 
The cost of drying film is not a major 
cost of the processing. However, a 
rapid drier should not require an 
extraordinarily high initial investment 
nor should its cost of operation be 
several times the operational cost of 
conventional driers. It should be con 
trolled easily and require a minimum o 
attention during operation. 

With these requirements in mind th 
merits of the possible methods can be 
considered. 

The basic differences in methods o 
drying have to do with the method o 
transferring the heat to the materia 
In drying film the three basic hea 
transfer methods can be used, namely 



88 



February 1953 Journal of the SMPTE Vol. 60 



conduction, radiation and convection. 
In order to use conduction the film 
must be in contact with a hot surface. 
Film driers have been built in which 
the support side of the film is brought 
in contact with hot, rotating drums 
while dry air is blown over the emulsion 
side. There are serious difficulties in 
obtaining uniform contact on the drums, 
but the principal objection to the system 
is that in order to maintain a reasonably 
small drum for high film speeds the drum 
temperatures must be above the soften- 
ing point of the support. Because the 
drums cannot be cooled quickly, acci- 
dental slowing or stopping of the 
machine will damage the film. Damage 
is also apt to result if the film dries too 
quickly while in contact with the drum. 

Radiation methods of heating the film 
are frequently used. These usually 
take the form of strip heaters or "infra- 
red" lamps. For the rapid drying of 
motion picture film there are serious 
objections to both of these. With any 
type of radiant-heat source the heat 
absorbed is a function of the film 
density. It has been shown that film 
subjected to infrared radiation will 
absorb more energy in the dark areas 
than in the light. Under accelerated 
drying this can lead to differential 
drying and film distortion is apt to 
result, particularly with silver-image 
films. 

The other factor which must be con- 
sidered is that the internal temperature 
of the film is a function of the radiation 
absorbed by it. The evaporation of the 
water from the emulsion will cool the 
surface and make its temperature ap- 
proach the wet-bulb temperature of the 
drying air, but the internal temperature 
of the film will be higher and may be 
above the critical limit of 240 F. This 
can be prevented, of course, by limiting 
the radiation to the value which the 
film will stand with the air conditions 
which are to be used while operating 
at the chosen film speed. These condi- 
tions could be determined experimen- 



tally for each film, but if the film speed 
accidentally decreased or the air tem- 
perature rose, the film would be 
damaged. 

Briefly, then, the principal objections 
to the use of infrared heat sources for 
motion picture film rapid driers are 
that careful control of film speed and 
drying air conditions are required to 
avoid damage either by overheating or 
differential drying. 

Internal heating can also be obtained 
by high-frequency currents. This is 
dielectric heating and is quite similar to 
other forms of radiant heating. This 
system can be self-regulating if properly 
installed. The energy absorbed by the 
film will be roughly proportional to the 
water content, if the correct wavelengths 
are used. In order to do this it would 
be necessary to use multiple stages of 
different wavelengths. In addition, air 
circulation is required to carry off the 
moisture. With this type of equipment 
very rapid drying can be obtained. 
However, because of the cost of the 
equipment for such an installation its 
use can probably not be justified except 
under very special circumstances. 

Convection, which is the third general 
method of heat transfer, seems to be the 
most suitable for this application. It 
has been shown that evaporating the 
moisture rapidly requires high-velocity 
air with its water-vapor pressure con- 
siderably lower than the water-vapor 
pressure at the film surface. This can, 
of course, be done by dehumidifying the 
air, but it can also be done by heating 
it, which is usually less expensive. 
Heating the air, then, serves the dual 
purpose of increasing the drying po- 
tential and supplying the heat required 
to balance the evaporation. The equip- 
ment required is simple and lends itself 
to inexpensive, automatic control. Fur- 
thermore, measurements of the wet- 
and dry-bulb temperatures will de- 
termine the actual emulsion tempera- 
tures during the constant-rate phase of 
drying and the maximum possible 



F. Dana Miller: Rapid Film Drying 



89 



Table I. Effect of Supply Air Humidity on Drying Time at Elevated Temperatures 



Supply air at 70 F 
Press, diff., 
% R.H. psi 



Pressure differences of heated supply air, psi 
125 F 150 F 200 F 



1 


20 


0.107 


0.290 


0.387 


0.583 


2 


90 


0.011 


.0.223 


0.322 


0.536 


Increase in dry- 












ing time, % 




87 


30 


20 


9 



temperature in the falling-rate phase. 
The emulsion temperature during the 
first phase is of importance in the drying 
of the few films which are rendered 
brittle at relatively low temperatures. 
In the latter phase, the maximum 
possible temperature is important in 
drying any film, so that the softening 
point of the support may be avoided. 
With the other methods of transferring 
heat to the film, these temperatures are 
frequently difficult to measure and it is 
therefore difficult to determine when 
they are operating within safe limits. 

Another advantage of using convection 
drying with high-temperature air is 
that fluctuations in the humidity of the 
supply air have little effect on the 
performance of the drier. This is 
illustrated in Table I, in which the 
vapor-pressure difference of 70 F inlet 
air at two different humidities is shown 
when heated to 125, 150 and 200 F. 
If all other conditions are maintained 
constant, the drying time will be in- 
versely proportional to the vapor- 
pressure difference. Line 3 shows how 
much the drying time would be in- 
creased at the various operating tem- 
peratures if the supply air humidity rises 
from 20% to 90%. Smaller humidity 
changes, which are more apt to occur, 
will have proportionally smaller effects 
on the drying time. The higher the 
temperature which is used the smaller 
the effect of humidity changes will be, 
and this is another incentive to use the 
highest temperature possible consistent 
with safe operation. 

It was consideration of these factors 
which lead to the decision that a drier 



using hot impinged air would generally 
be the most satisfactory for rapidly 
drying motion picture film. No doubt 
special situations will arise in which 
other methods will be indicated. One 
of the objects of this investigation, 
however, was to investigate the method 
which seemed to have the widest possible 
application in the industry. 

Figure 3 illustrates the equipment 
used. The film is fed from the supply 
reel through an air squeegee, over the 
holdback sprocket, between the air 
plenum chambers, and on to the take- 
up. The drying air is supplied by a 
turbine-type blower capable of supply- 
ing 85 cfm at 6-in. water pressure. 
Two 1300-w electric heaters in the 
discharge duct from the blower heat the 
air before it enters the plenum chambers. 
The plenums are 3 ft long with 0.080 
in. diameter holes facing the film. There 
are 128 holes impinging air on the sup- 
port side of the film and 256 holes 
impinging air on the emulsion side of 
the film. 

The drive is by means of a d-c motor 
controlled by a rheostat, through a gear 
reducer to the holdback sprocket. The 
take-up reel is driven by a friction belt 
take-off from the gear reducer. A 
rheostat on the blower, as well as 
orifices in the inlet, is used to control 
the air volume and therefore the velocity 
of impingement. The temperature of 
the air is regulated by varying the 
number of heaters used. The air from 
the plenum chambers, after impinging 
on the film, is discharged from the 
cabinet through the entrance and exit 
slots for the film. The plenum chambers 



February 1953 Journal of the SMPTE Vol.60 




Fig. 3. Experimental Impingement Drier for 35mm Film. 
The upper righthand corner is an enlargement of a section of the drying chamber. 



are 1 in. apart and guide bars are used 
to insure that the film is maintained 
midway between them. 

Figure 4 is a sketch of the squeegee. 
This is a Capstaff 5 type of squeegee in 
which the lower rollers form a highly 
efficient air knife. 



The experiments were made with film 
which had been previously processed 
and dried. This was re wetted by 
winding through 70 F water three times 
in a 20-min period. In the interval 
between windings the roll of film was 
submerged in water. Comparison of 



F. Dana Miller: Rapid Film Drying 



91 





GAP 0005* MORE 

THAN FILM THICKNESS 



Fig. 4. Squeegee. 

the drying of film rewetted in this 
manner with film which was dried on 
the equipment immediately after proc- 
essing showed that there was no 
measurable difference in the required 
drying times. It is extremely fortunate 
that this was the case because the 
complication of having to process the 
film as it was being used would have 
made an extensive survey impractical. 

In making an experiment the air 
temperature and pressure were adjusted 
and the machine allowed to run until 
it had reached constant conditions. 
Film was then attached to the leader, 
threaded through the machine the 
supply roll still being submerged in 
the tank of water and the machine 
drive started. It was allowed to run 
at the chosen film speed until steady 
conditions prevailed. Samples of the 
dry film were cut from the film as it 
emerged from the end of the dry cabinet 
and placed in stoppered bottles. Wet 
film samples were then cut from the 
film just beyond the holdback sprocket 
after it had emerged from the squeegee 
and before it entered the drying cabinet. 
The moisture content of the film was 
determined by accurately weighing the 
samples in the bottles before and after 
being subjected to an evacuation pro- 
cedure. The difference between these 



two readings accurately determined the 
moisture loss of the film passing through 
the drier. For each set of air conditions 
investigated, the effect of drying time 
was explored by making experiments 
at a < series of film speeds, thereby ob- 
taining different drying times. 

In Fig. 5 is shown the performance of 
the machine on Eastman Fine Grain 
Release Positive Safety Film Type 5302. 
In this plot the ordinate is the moisture 
content of the film and the abscissa is 
the drying time. The impingement 
velocities used were 2,000 and 4,000 
fpm and at each of these, temperatures 
of 125, 150 and 200 F were used. The 
dry point, indicated as a horizontal line 
on the graph at the 2.5% moisture line, 
was arbitrarily chosen and is the point 
at which the film has the same moisture 
content as it would have if it were in 
equilibrium with air at 70 F and 50% 
R.H. To illustrate the margin of 
safety, the 60% and 30% R.H. levels 
are also indicated. Under the most 
severe drying conditions used it can be 
seen that required drying times of 10 
sec were obtained and the longest 
required drying time was somewhat 
more than 42 sec at 125 F and 2,000 
fpm air velocity. The moisture ab- 
sorption characteristics of emulsion vary 
to a considerable extent. This emulsion 
is quite hard and as a result the film 
absorbs only about 15 to 19% moisture. 
The performance of the drier on films 
of relatively high moisture capacity is 
shown in Fig. 6. This is Eastman Plus 
X Panchromatic Negative Safety Film, 
Type 5231, which absorbs approximately 
35% moisture. With this film the 
minimum required drying time was 16 
sec and the maximum was more than 
1 min. 

General Curve Shape 

Interpreting these curves in terms of 
the classical constant- and falling-rate 
phase of drying, it might be said that 
the early part of the drying approxi- 
mates the constant-rate phase, then 



92 



February 1953 Journal of the SMPTE Vol. 60 




10 20 30 40 50 
DRYING TIME- SECONDS 



Fig. 5. Impingement Drying of Eastman Curve 

Fine Grain Release Positive Safety Film, j 

Type 5302. 2 

For satisfactory properties, final moisture 3 

should fall in shaded area. Required 4 

drying time is determined by intersection 5 

of curves with dotted line at 50% R.H. 6 



Air Temp. : 

125 
125 
150 
150 
200 
200 



Air Velocity, fpm 

2,000 
4,000 
2,000 
4,000 
2,000 
4,000 



the curve shape reverses, indicating a 
transition, and the final part of the 
curves are in the falling-rate phase. 
Calculation of these data has shown 
that the first part of the drying cycle is 
not actually at constant rate. This is 
not surprising because drying of the 
support is limited by the rate of diffusion 
of water through the support, which is 
a falling-rate type of drying. At the 
same time, the emulsion is probably 
being dried at almost constant rate. 



Because the amount of water being 
evaporated from the emulsion during 
this phase is so much greater than that 
evaporated from the support, the combi- 
nation of the two approximates con- 
stant-rate drying. As the drying of the 
emulsion progresses, the rate at which 
water can migrate to the surface of 
evaporation becomes less than the rate 
at which it can be evaporated and the 
drying of the emulsion as well as the 
support is done at a falling rate. 



F. Dana Miller: Rapid Film Drying 



93 







20 30 40 50 

DRYING TIME -SECONDS 



60 



Fig. 6. Impingement Drying of Eastman Curve Air Temp., F Air Velocity, fpm 



Plus X Panchromatic Negative Safety 

Film, Type 5231. 

For satisfactory properties, final moisture 
should fall in shaded area. Required 
drying tune is determined by intersection 
of curves with dotted line at 50% R.H. 



125 
125 
150 
150 
200 
200 



2,000 
4,000 
2,000 
4,000 
2,000 
4,000 



It is not the purpose of this paper to 
make a detailed mathematical analysis 
of the drier performance. However, a 
few general conclusions do seem to be 
warranted and may be helpful to others 
who may wish to construct this type of 
equipment. 

Effect of Temperature on Drying 
Time 

With any constant velocity of drying 
air up to 6,000 fpm, raising the air 



temperature from 125 to 150 F will 
reduce the required drying time about 
50%. By raising the temperature from 
150 to 200 F, the reduction in required 
drying time will be between 30% and 
50%. These data are derived not 
only from that shown in Figs. 5 and 6, 
but also from experiments on a variety 
of other types of films. It was noted, 
and the two examples given are typical, 
that for nearly all positive films the 
reduction in required drying time by 



94 



February 1953 Journal of the SMPTE Vol. 60 




10 



20 30 
DRYING TIME 



40 50 
SECONDS 



Fig. 7. EflFect of Air Velocity on the Drying of Eastman Super XX. Panchromatic 
Negative Safety Film, Type 5232, at 125 F Air Temperature. 



Curve 
1 
2 
3 



Air Velocity, fpm 
2,000 
4,000 
6,000 



increasing the temperature from 125 
to 150 F was about the same as the 
reduction obtained by increasing the 
temperature from 1 50 to 200 F, whereas 
with negative films, increasing the tem- 
perature in the latter range usually 
resulted in less than a 50% reduction in 
required drying time. 

Effect of Air Velocity 

At low air temperatures increasing the 
air velocity from 2,000 to 4,000 fpm has 



a negligible effect on the required drying 
time. This was true for both positive 
and negative type films. At 150 or 
200 F the same change in air velocity 
resulted in decreasing the required 
drying time from 16% to 45%, with the 
greater reduction being at the higher 
temperature. 

These remarks should not be con- 
strued as meaning that there is a simple 
relationship between increase in air 
velocity and decrease in required drying 



F. Dana Miller: Rapid Film Drying 



95 




10 



20 30 40 50 
DRYING TIME- SECONDS 



Fig. 8. Effect of Air Velocity on the Drying of Eastman Super XX Panchromatic 
Negative Safety Film, Type 5232, at 150 F Air Temperature. 



Curve 
1 
2 
3 



Air Velocity, fpm 
2,000 
4,000 
6,000 



time. This is definitely not the case. 
It was found that with any particular 
air temperature there is a maximum air 
velocity beyond which the decrease in 
required drying time is negligible. 
Eastman Super XX Panchromatic Nega- 
tive Safety Film, Type 5232, was dried 
at 125 F with 2,000-, 4,000- and 6,000- 
fpm air velocities. These same velocities 
were used with 150 F air and 200 F air. 
The results are shown in Figs. 7. 8 and 
9, respectively. The actual dry point 



of this film is 3.1%, but since not all 
the experiments were carried to this 
point comparison at the 4% level 
corresponding to 60% R.H. will suffice. 
At 125 F there is a reduction of 35% 
to 40% in the required drying time by 
increasing the air velocity from 2.000 
to 4,000 fpm, but when the velocity is 
raised to 6,000 fpm no measurable 
difference in required drying time is 
found. This is also true with air 
temperatures of 150 F as shown in 



February 1953 Journal of the SMPTE Vol. 60 




20 30 40 50 
DRYING TIME- SECONDS 



60 



Fig. 9. Effect of Air Velocity on the Drying of Eastman Super XX Panchromatic 
Negative Safety Film, Type 5232, at 200 F Air Temperature. 



Curve 
1 
2 
3 



Air Velocity, fpm 
2,000 
4,000 
6,000 



Fig. 8. No doubt there was actually 
some improvement in drying when the 
air velocity was increased from 4,000 
to 6,000 fpm, but it was so slight that 
it could not be detected in these experi- 
ments. Figure 9, however, shows that 
at 200 F increasing the air velocity 
through this range reduces the required 
drying time from 17 sec to 14^ seconds 
or about 15%. Apparently, at this 
temperature, somewhat faster drying 
time could be obtained at velocities in 



excess of 6,000 fpm but the experiments 
indicate that at some point, probably 
about 8,000 fpm, practical minimum 
required drying time would be attained. 
Figures 5, 6, 7, 8 and 9 show that 
the velocity at which the required drying 
time reaches a minimum is a function 
of the drying air temperature. It is 
also believed to be dependent on the 
particular arrangement of the machine. 
The size, number, and spacing of the 
air jets as well as the passage of the air 



F. Dana Miller: Rapid Film Drying 



97 



II 




III* 



i- 

1 

X - 

a *. 




*f o co o CN vot^-t-ioo i CN ^- \o o o m \o oo r oo o cs m oo 

cO cO cO *<* cO cO cO cO cO cO cO cO cO CN cO CO CO CO CO -^- CO f> ^- CO 



cO cO CO CN r~ cO CN CO CO CO CO cO CO CO 

p o o o p p p p p p p p p p 
ooooo oooo oooo d 



p p p p p p <- 
o ooooo ooo 



ooooo oooo oooo odd do od d 



T-> ^- ,-1 CO 00 v* CO CN IT) \O CN <* Tf CO CN IT) m ^-i CN CO iTi CN CN CO CO 

p p p p p p p p p p p ^- p pop p p p p p p p p p 

odddd dodo" dddd o'dd ddddd dodo 

I + + + + ++ I + + + + + + + + +1 I I + + I + + 



m o CN r~~ 

0\" CN O O '-- 



- r CN -r* 



CN 00 C\ CN 



in 00 * m t~^ CN CN CN t~- *-i 
m CN CN T-I CO T-I T-> ^ CN CN 



i 

00 



^ CN -<j- Tf CN ^^CN^- ^^^ "t CN 




.s 

o 



OH 
fa 





February 1953 Journal of the SMPTE Vol. 60 



over the film to the exhaust must affect 
this value as well. 

Effect of Squeegeeing on Drying Times 

The majority of this work was done 
with the squeegee adjusted to a constant 
pressure of 3 psi. This pressure was 
selected after a series of experiments 
showed that this was the minimum pres- 
sure which would insure that all the 
surface-water drops would be removed 
at the film speeds used. 

Under these conditions the drying 
effect of the squeegee varied with the 
film speed; therefore, the moisture 
content of the film entering the cabinet 
was a function of the film speed and was 
different for each point on a drying 
curve. In the case of Fine Grain Re- 
lease Positive, the initial moisture con- 
tent of the film dropped as much as 
3.5% moisture as the film speed was 
decreased from the maximum to the 
minimum drying during the determi- 
nations of the points on a drying curve. 
The same amount of change in the 
initial moisture occurred in determining 
drying curves of negative films absorbing 
as much as 45% moisture. 

The required drying times found in 
Figs. 5 through 9 were obtained at lower 
film speeds and therefore at the lower 
moisture content of the entering film. 
It would be expected that the required 
drying times would be greater if the 
initial film moisture was higher. In- 
vestigation of this point showed that 
the maximum difference in the required 
drying times was of the order of 10% 
and this was found only with films 
absorbing less than 20% moisture, such 
as Fine Grain Release Positive. Drying 
curves at constant initial moisture were 
obtained on Super XX Panchromatic 
Negative Film, which absorbs 40% 
moisture, but comparison of these 
results with those in which the initial 
moisture varied showed no difference in 
required drying time. 

If the performance data of this equip- 
ment are used as the basis of design for 



commercial equipment, it should be 
assumed that the commercial squeegee 
would be adjusted for a minimum of 
drying and therefore the required drying 
times given for Fine Grain Release 
Positive should be increased by 10%. 
The required drying times of the other 
films are unaffected by this slight change 
in initial moisture and do not require this 
correction. 

Effect on Film Properties 

In general the physical properties of 
the films dried under these conditions 
differed very little from the properties 
of the same films dried on conventional 
machines. The author wishes to stress 
that these remarks apply only to the 
films actually tested and that it would be 
a mistake to assume that all films would 
react similarly. Careful measurements 
of the brittleness, curl level, humidity 
curl amplitude and distortion were 
made on all of the films which were 
used in the experiments. A comparison 
of these values with typical values of 
normally dried film showed no out- 
standing differences. The films used 
in the investigation and the measure- 
ments made are listed in Table II. In 
drying the films for these measurements, 
the machine speed was adjusted so that 
the film would be dried to approxi- 
mately a 50% R.H. equilibrium mois- 
ture content. In addition, some samples 
were purposely overdried, particularly 
at the higher air temperatures in order 
to show any effects of overdrying. 

When the films were not overdried, 
the rapid drying had little effect on 
either the curl level or the humidity 
curl amplitudes. In the case of the Fine 
Grain Release Positive there was some 
lowering of both of these values. Over- 
drying did not seriously affect either 
of these properties, although there is a 
slight indication that it may lower the 
humidity curl amplitude of some of the 
films. 

The effect on the brittleness, as indi- 
cated by the Vise Brittleness and Pfund 



F. Dana Miller: Rapid Film Drying 



99 



Fold Tests,* was negligible for all the 
films except the Super XX Panchro- 
matic Negative. Overdrying this film 
increased the brittleness. However, 
when the film was properly dried, it was 
not brittle. 

An examination of the various films 
for distortion showed that nearly all 
of them were fluted. This fluting 
seemed to be more severe on the films 
which had been dried at temperatures 
above 180 F. Poor alignment in the 
experimental machine may have been 
responsible for the distortion. It also 
may have been caused by the softening 
of the hard rubber spools at these 
temperatures with the result that the 
films tended to climb the flanges rather 
than to be guided by them. It is 
possible that the distortion could be 
eliminated in a well-built machine. 
However, our experience illustrates how 
easily the film can be distorted at these 
temperatures, and, therefore, the neces- 
sity of extremely good film transport 
mechanism in machines of this type. 

Attempts to evaluate the tendency for 
the rapidly dried film to go "in-and-out 
of focus" 6 have not been very successful 
to date. It can be said that none of the 
rapidly dried films equalled the per- 
formance of the best of the normally 
dried films with which they were com- 
pared. On the other hand, little differ- 
ence could be found between the average 
performance of all of the rapidly dried 



* The Vise Brittleness Test consists of 
breaking a loop of film, emulsion side out, 
between the jaws of a vise closed at uni- 
form speed. The distance between the jaws 
at the instant of film failure is the value 
given as vise brittleness. This is usually 
done in a room held at 70 F and 10% 
R.H. after the film has been equilibrated 
to these conditions. 

The Pfund Fold Test consists of the fold- 
ing of a short free loop of film alternately 
emulsion in and out. A fold is considered 
as a complete cycle. This test is also run 
at 70 F and 10% R.H., with the film 
equilibrated to those conditions. 



films when compared with the average 
performance of all of the normally 
dried films used in the test. 

The "in-and-out of focus" test used is 
rather severe and does not simulate 
trade^ conditions. It consisted of the 
continuous projection of a 45-ft loop of 
film by an arc projector operated at a 
light intensity approximately equal to 
the maximum in present theaters. The 
arc was a condenser type, burning a 
13.6-mm positive at 170 amp, and 
equipped with an //2.0 quartz con- 
denser set to deliver a mean net radiant 
flux to the film of about 0.5 watts/sq m. 
The image is projected with an //2.0 
projection lens, and an observer sets 
the focus each time the start of the loop 
passes through the projector and records 
the focus position as indicated by a dial 
indicator. He also records the appear- 
ance of the image. The test continues 
until the image is steady and the focus 
becomes constant or until the loop has 
made 25 cycles. Under these conditions 
film will almost always go "in-and-out 
of focus," but the more cycles a loop 
of film will travel before this occurs the 
less the chance that it would give trouble 
in actual theater use. 

High-intensity projection of a 1000-ft 
roll of rapidly dried film in a trade use 
test did not cause "in-and-out of focus." 
This roll was projected four times per 
day for forty projections at the end of 
which time the focus position had 
become constant. 

Further work is being done on this 
problem but at present it must be con- 
cluded that the "in-and-out of focus" 
tendency of rapidly dried film may be 
somewhat greater than with normally 
dried film. If commercial installations 
of this type of drier are made, the 
product should be carefully checked for 
this defect and procedures established 
which will minimize the tendency. 

Projection life of the Fine Grain 
Release Positive Film rapidly dried was 
the same as for conventionally dried 
product as were also the tear values. 



100 



February 1953 Journal of the SMPTE Vol.60 



Table III. Experimental Impingement Drier Air and Heat Requirements 



Impingement 
velocity, fpm 


Air vol., 
cfm 


Air 
in 


press., 
H 2 


125 


F 


Heat required, 
150 F 


Btu/min 
200 


F 


2,000 
4,000 
6,000 
Effective 


28 
57 
85 
film path 43 




2 
6 
in. 


.69 

.7 
.25 


28. 

57. 
85. 


2 

2 

4 


37.4 
76.3 
113.5 


56 
115 
171 


.8 
.5 
.8 



Sensitometric Properties of Fine 
Grain Release Positive Safety Film 

D. R. White 7 pointed out that drying 
conditions can change density and 
contrast depending on the humidity 
and temperature of the air used. An 
experiment was made by printing a 
variable-area and a variable-density 
sound track on two parts of the same 
roll of Fine Grain Release Positive, 
processing both in the normal manner 
and drying half of each part in the 
normal manner, while the remainders 
were dried in 6 sec with air at 210 F. 
The rapid drying increased the optimum 
density of the variable-density sound 
track by 0.03 and decreased the density 
of the variable-area sound track by 0.10. 
The density shift in the former case is 
not considered significant, but the 
latter is significant and would require 
compensation if repeated tests showed it 
to be real. 

Rapid Drying of Rapidly Processed Film 

While it is not the purpose of this 
paper to discuss the drying of rapidly 
processed films, it is felt that some 
mention of the problem should be made 
because of the general interest in the 
subject as a result of discussions of it in 
the current literature. 

The rapid drying of rapidly processed 
film which was discussed by Katz 8 is a 
problem somewhat different from the 
rapid drying of normally processed film. 
Because of the short immersion time, 
only the emulsion absorbs an appreciable 
amount of water and even this ab- 
sorption may be limited by the process- 
ing. 



To illustrate the effect that this 
method of processing has on required 
drying time, it was found that rapidly 
processed Fine Grain Release Positive 
Film could be dried on this equipment in 
3.6 sec with 200 F air at a velocity of 
4,000 fpm and in 3.0 sec at 6,000-fpm 
air velocity. This is approximately 
one -third of the time required to dry the 
same film, under the same conditions 
after normal processing. 

Because each emulsion as well as each 
film support has its own moisture 
absorption characteristics, a comparison 
of this drying time with the drying time 
of another machine using a different 
film and process has very little meaning. 

Power Consumption 

In order to convey some idea of the 
economics of this type of drying, the air 
and heat requirements of the experi- 
mental drier are given in Table III. 
No power measurements were made 
on the blower because it was realized 
that the power for full-scale equipment 
could be much more accurately de- 
termined from the manufacturers' cata- 
logs on the basis of the pressure and 
volume of air required. The heat re- 
quirements are based on using 100% 
fresh air at a temperature of 70 F and 
50% R.H. and assuming that there 
would be a 10% loss of heat in the 
system. 

With these data it is possible to 
calculate the general requirements of a 
production drier and obtain some 
conception of its physical size, as well 
as power requirements. For example, 
assume that an impingement drier is 



F. Dana Miller: Rapid Film Drying 



101 




Fig. 10. Arrangement of an Impinge- 
ment Drier Cabinet for Drying Fine 
Grain Release Positive Film at 300 fpm. 

to be used for drying Fine Grain Release 
Positive Safety Film at 300 fpm with 
200 F air at a velocity of 4,000 fpm. 
From Fig. 5, the required drying time 
for this film for these conditions is 10 
sec. To this must be added 10% to 
compensate for the low initial moisture 
content of the experiment, making the 
actual required drying time 11 sec. 
This means the film path in the drier 
must be 11/60 X 300 = 55 ft. This is 
approximately 15.4 times as large as 
the pilot-plant machine. Therefore, 
the total air requirement is 900 cfm and 
the heat required is 1800 Btu/min. 
The pressure required to deliver air 
at this velocity through the orifices is 
2.7 in. of water. Any fan similar to a 



Glarage size 1 type W single-width 
fan running at about 1800 rpm and 
powered with a 1-hp motor would be 
suitable for this service. 

Figure 10 is a sketch to illustrate the 
approximate size of the drier cabinet 
and showing some of the features which 
would be desirable in such an installa- 
tion. 

In this arrangement the machine 
would have a single rack of film 5 ft 
high with 6 strands on it. Three-inch 
diameter film spools could be used. 
Backing rollers would be required so 
that the maximum unsupported film 
length would be 20 in. The air would 
be supplied from a large duct at the rear 
of the film cabinet. The three sections 
of the center plenum chamber would 
be rigidly attached to this duct and 
open into it. The two outside plenum 
chambers impinging on the emulsion 
surface would also open into the duct; 
however, they could be hinged on the 
duct so that they could be swung aside 
to facilitate inspection of the film or 
threading up. Inside the air duct at 
the openings to the side plenums, flap- 
type dampers could be provided which 
would automatically close off the air 
flow to the plenum chambers when they 
were swung aside. 

Attached to the outside plenums, 
across the vertical edges, glass panels 
are shown and these together with the 
metal flaps attached to the bottoms 
would form a cabinet when the plenums 
were in their operating position. The 
top of the cabinet would be formed by 
the fixed exhaust duct which should be 
piped outside the building. 

It would probably be desirable to 
install an automatic damper in the main 
air supply to shut off the air to the 
plenums in case of accidental stops 
while film is in the cabinet. The space 
occupied by the machine would be 
approximately 1 X 2 ft without fan or 
heaters. 

Since this sketch and description are 
not intended to be a finished design, 



102 



February 1953 Journal of the SMPTE Vol. 60 



no attempt has been made to show the 
equipment in detail and necessary items 
such as film drive, take-up mechanism, 
fan and controls have been omitted. 
There are many ways in which this 
length of thread-up and air system 
could be arranged and this is not pre- 
sented as the best. It does serve to 
point out graphically how compact the 
equipment can be. 

The cost of the equipment would 
probably be less than for conventional 
driers of similar capacity. The power 
requirements are actually less than for 
many conventional driers of less capacity. 

Remarks 

In contemplating the installation of 
impingement driers for commercial work 
there are several factors which should 
be considered. 

The first of these is the problem of 
squeegeeing. In rapid drying the re- 
moval of all surface water is essential 
if water spots are to be avoided. 

It should also be noted that once the 
drier is built experiments will have to 
be made with each film which it is 
intended to dry, in order to determine 
the operating conditions which will 
satisfactorily dry the film at the design 
film speed. The operator's judgment 
will not be adequate in this case. The 
curl level of rapidly dried film, as it 
emerges from the cabinet, is not the 
same as it will be after the support and 
the emulsion have equilibrated. Fur- 
thermore, the film may feel dry and warm 
as it emerges even though insufficiently 
dry. This film may become tacky after 
it is rolled up and the base and emulsion 
have equilibrated. The proper condi- 
tions can be found by trial and error 
or by making moisture analyses of the 
film dried under different air conditions. 
Once these conditions are established 
the acceptable moisture range provides 
liberal margins of safety with respect 
to either air temperature or velocity. 
It is also possible that one such set of 
conditions will be satisfactory for drying 



a variety of films. For example, it was 
found that with a drying time of 12 
sec and air at 200 F and 4,000-fpm 
velocity, Fine Grain Release Positive, 
as well as both types of Sound Record- 
ing film would be dried properly. 
However, this could not be determined 
by observation of the film. It is be- 
lieved that this is the principal dis- 
advantage of this type of drying. 

The purpose in presenting these 
data has not been to encourage the use 
of this type of drying for any and all 
applications. The experiments do in- 
dicate that the method is applicable to 
a variety of films. It is known, however, 
that 16mm Kodachrome will become 
extremely brittle if impingement dried 
at temperatures higher than 125 F. 
It is possible that other films will have 
similar limitations. It is therefore ur- 
gently suggested that before designing 
or installing this type of equipment, the 
manufacturers of the various types of 
films which are to be processed in the 
machine be consulted. 

Conclusions 

This survey has shown that many 
motion picture films can be safely dried 
with hot impinged air. Drying times 
ranging from 10 to 30 sec with air 
temperatures as high as 200 F did not 
have undesirable effects on the physical 
properties of the films tested. There is 
an indication that the film density is 
affected by the treatment but the 
amount of change is small. 

Consideration of several methods 
which might be used for rapidly drying 
motion picture film seem to indicate 
that the most suitable method for com- 
mercial application is by means of high- 
velocity hot air. 

While the method has been used 
successfully in experimental tests on a 
wide variety of films, it would be wise 
for anyone contemplating the installa- 
tion of such equipment to determine 
first its effects on each of the types of 
film which he intends to use in it. 



F. Dana Miller: Rapid Film Drying 



103 



Acknowledgments 

The writer wishes to express his 
appreciation to Dr. C. R. Fordyce for 
his extremely helpful guidance in this 
work. The contributions and sugges- 
tions of numerous members of the 
Manufacturing Experiments Division 
and the Kodak Research Laboratories 
are also sincerely appreciated. He is 
particularly indebted to Warren E. 
Duerr who carried out a large part of 
the work. 

The subject matter contained herein 
is for the reader's information only, 
and none of the statements contained 
herein should be considered as a recom- 
mendation for the manufacture or use 
of any substance, apparatus or method 
in violation of any patents now in force 
or which may issue in the future. 



References 

1. E. K. Carver, R. H. Talbot and H. A. 
Loomis, "Film distortions and their 
effect upon projection quality," Jour. 
SMPE, 41: 88-93, July 1943. 

2. C. E. Ives and G. Kunz, "Simplification 
of'motion picture processing methods," 
Jour. SMPE, 55: 3-26, July 1950. 

3. F. J. Kolb, Report of a survey of New 
York and Hollywood laboratories (un- 
published report, Eastman Kodak Co.). 

4. J. H. Perry, Chemical Engineers Hand- 
book, Third Edition, p. 803, 1950. 

5. J. G. Capstaff, Eastman Kodak Com- 
pany, U. S. Pat. No. 2,289,753 (1942). 

6. E. K. Carver, R. H. Talbot and H. A. 
Loomis, "Effect of high intensity arcs 
upon 35mm film projection," Jour. 
SMPE, 41: 69-87, July 1943. 

7. D. R. White, "Drying conditions and 
photographic density," Jour. SMPE, 19: 
340-345, Oct. 1932. 

8. L. Katz, "Drying film by turbulent 
air," Jour. SMPTE, 56: 264-279, Mar. 
1951. 



104 



February 1953 Journal of the SMPTE Vol. 60 



Further Experiments in High-Speed 
Processing Using Turbulent Fluids 

By LEONHARD KATZ and WILLIAM F. ESTHIMER 



The results of an investigation into the effects of turbulence in the rapid 
processing of photographic film under fixed conditions are described. For 
test purposes a turbulent processing machine was constructed permitting 
rapid switching of fluids and evaluation of resulting reactions. 



A RESULT of a number of investiga- 
tions made involving mass transfer under 
conditions of turbulent flow, the possi- 
bility of increasing the speed of develop- 
ment of photographic film by means of 
turbulent liquids and gases was taken 
under consideration during November 
1949. As a result of some theoretical 
work, an initial machine was constructed 
as shown in Fig. 1 for the purpose of ex- 
perimenting with small strips of motion 
picture film subjected to turbulent de- 
veloper. These experiments were com- 
pleted in April 1950 and have been de- 
scribed in Ref. 1. 

The experiments indicated that at 
room temperature turbulent developing 
fluid acted between 2 and 3| times as 
fast as normally agitated developing 
fluids, using standard developers and 
positive-type film. Some experiments 

Presented on October 8, 1952, at the 
Society's Convention at Washington, D.C., 
by Leonhard Katz, Woburn Engineering 
Co., 19 Ward St., Woburn, Mass., who 
read the paper, and William F. Esthimer, 
Raytheon Mfg. Co., Waltham 54, Mass. 
This is a report of research carried out at 
the Raytheon Mfg. Co., Waltham 54, Mass. 



were also performed using different 
films and mono-bath solutions (i.e., 
solutions for the simultaneous developing 
and fixing of an emulsion) in cooperation 
with the Optical Research Laboratory, 
Boston University, during April 1950. 
It was observed that the time factor 
varied between 2 and 3j and was 
apparently influenced by variations in 
the thickness of the emulsion and the 
type of film used. It was also observed 
that the normal equilibrium between 
developing and fixing solutions in the 
mono-bath was upset under conditions of 
turbulent flow so that the relative rate 
of developing and fixing was changed. 

Although the initial experiments were 
rather incomplete it was observed that 
the time factor increased as the Reynolds 
number was increased and no apparent 
leveling off of the curve was observed 
within the limited region of the experi- 
ments (the maximum Reynolds number 
was 90,000). In addition, it was 
observed that with thin emulsions the 
time factor (over normal agitation 
developing) increased to approximately 
5j at slightly elevated temperatures. 
It appeared that the application of 



February 1953 Journal of the SMPTE Vol. 60 



105 



Symbols, Definitions, Dimensions and Values 



A = area normal to heat or mass trans- k 

fer, sq ft Pr 

B = thickness of stagnant layer, ft Re 

C = concentration, % Sc 

Cp = specific heat, Btu/lbF T 

D = diameter, ft V 

E a = rate of mass transfer, Ib/hr 5 

g = acceleration of gravity, ft/sec j n 

h = heat transfer coefficient Btu/hr- p 

sq ft F 



thermal conductivity 
Prandtl number, C p n/k 
Reynolds number, 
Schmidt number, up/8 
temperature, R 
velocity, fpm 
diffusivity, sq ft/hr 
viscosity, Ib/ft hr 
density, Ib/cu ft 
partial pressure, atm 



turbulent fluids might well provide an 
important time gain in the rapid proc- 
essing of photographic film and that 
further investigation would the justified. 
It was, therefore, decided to embark on 
a research program in which an effort 
would be made to isolate the results 
which were solely attributable to the 
effect of turbulence so that a proper 
evaluation could be made. 

Scope of the Investigation 

Although it was realized from the 
beginning that the effects of turbulent 
flow would be more useful if they were 
combined with other changes, such as 
increased concentration of developer, 
different emulsions, different tempera- 
tures, etc., it was decided to limit this 
first investigation solely to the investiga- 
tion of the effects of turbulence. Conse- 
quently, the following standards were 
established : 

1. The film to be used was Kodak 
Super-XX Aerographic Film. 

2. The developer to be used was 
Kodak Developer D-19.* 

3. The temperature at which the film 
was to be processed was to be maintained 
at 68 F (66 F to 70 F). 

4. The fixer to be used was Kodak 
Fixing Bath F-6.* 

5. The developing time under turbu- 



lent conditions was the time measured 
between the initiation of turbulent flow 
and the cessation of turbulent flow. 

6. The fixing time was determined 
as the time to clear the film completely 
of opalescence. 

7. Washing time was determined by 
silver nitrate tests.** 

8. Small samples of film approxi- 
mately l| in. wide by 6j in. long, on 
which a photographic wedge had been 
exposed, were used in a stationary 
mount. 

Theory 

The treatment of photographic emul- 
sion by means of various chemicals 
to perform the functions of developing, 
fixing, washing and drying and other 
basic functions can, in general, be 
separated into two basic phenomena 
as follows: 

1 . The chemical reaction taking place 
between the particles contained in the 
photographic emulsion and the particles 
introduced by external means. 

2. The diffusion by which these 
chemicals are transported from the sur- 
rounding atmosphere into and out of 
the gelatin to permit the accomplish- 
ment of the chemical reaction. 

Photographic film can be considered 
as consisting of a base which absorbs 
practically no fluids, and a gelatin 



* See Photo Lab Index, Henry M. Lester, 
Sect. 6, Morgan & Lester, New York, 
12th ed., 1952. 



** The Theory of the Photographic Process, 
C. E. K. Mees, Macmillan, New York, 
1942, p. 532. 



106 



February 1953 Journal of the SMPTE Vol. 60 



DEVELOPING CHAMBER 



P--J TO FAUCET 




MODEL Z EASTERN PUMP 



Fig. 1. Model developing test set-up. 



layer which can absorb a large amount 
of fluids. The processing of film requires 
that the molecules of the processing 
fluids diffuse into the gelatin layer and 
out of the gelatin layer depending on 
whether the process is one of developing, 
fixing, washing, or drying. This transfer 
of fluids into and out of the gelatin layer 
serves only the purpose of transporting 
a small amount of chemicals which can 
then react with other substances already 
present in the emulsion. It can be 
seen that if this transfer rate of fluids 
can be increased, the speed of chemical 
reaction will, in general, also be in- 
creased inasmuch as the rate of chemical 
reaction will be dependent on the 
number of chemical particles which are 
carried into or out of the emulsion by the 
transporting fluid. 

For instance, a developing fluid will 
consist largely of water containing vari- 
ous chemical substances. Upon sub- 
mersion of a film in the developer, the 
fluid will penetrate the gelatin, taking 
the chemical substances with it. The 
actual developing process inside the 
gelatin is then governed by statistical 
rates which determine the total number 



of molecules of the developing agent 
which come in contact with the various 
molecules of silver halides which are in 
the gelatin. As the developing process 
takes place, a concentration gradient 
will be established in the emulsion such 
that the amount of active ingredients, 
i.e., unused developer, decreases at 
greater depth in the emulsion. Given 
a specific emulsion and a specific chemi- 
cal formula to be used for the developing, 
there are a number of factors which 
can be used for increasing the velocity 
of development as follows: 

1. The concentration of the developer, 
i.e., the number of available molecules, 
can be greatly increased. As a result 
of this, the probability of a silver halide 
molecule making contact with the 
developer molecules is greatly increased. 

2. The velocity of the developer 
molecules may be considerably increased 
by an increase in the temperature of 
the developer, which will so increase 
the agitation of the molecules that the 
probability of collision is again greatly 
increased. 

3. The rate of diffusion through the 
gelatin may be increased, thereby re- 



Katz and Esthimer: Turbulent Fluid Processing 



107 



during the concentration gradient in 
the gelatin. Consequently, the concen- 
tration of effective developer inside the 
emulsion is enormously increased, which 
again increases the probability of colli- 
sion. 

It should be noted that the three 
variables mentioned are to a great 
extent independent variables. 

Basic Theory of Mass Transfer 1 

In order to study the problems in- 
volved in diffusion more closely, it is 
well to consider first the general theory 
of mass transfer between a solid sub- 
stance and a liquid or gaseous sur- 
rounding. From the general theory 
it will then become apparent which 
parameters can be varied to obtain the 
maximum rate of mass transfer and so 
obtain the maximum speed of processing. 

It has long been recognized that heat 
and mass transfer problems are similar 
in nature and that the same theory will 
apply to both problems. The heat and 
mass transfer between a liquid and solid 
will depend on a number of variables 
as follows: 

1 . The area of contact through which 
heat or mass transfer takes place. 

2. The driving force, i.e., the tempera- 
ture difference in case of heat transfer 
or concentration difference in case of 
mass transfer. 

3. The resistance to mass or heat 
transfer which is proportional to the 
thickness of a stagnant layer existing at 
the dividing surface between the two 
media. 

The first two variables have usually 
been well recognized, and it is the usual 
practice to increase heat or mass transfer 
by increasing the surface area or increas- 
ing the driving force, i.e., raising the 
temperature. 

The third variable takes into account 
the resistance to heat or mass transfer 
in the stagnant layer. This will be 
especially important where the stagnant 
layer is a controlling factor, as is the 



case in the developing and drying of 
photographic film. The stagnant layer 
as used in the following discussions can 
be considered as a relatively non-moving 
layer of liquid or gas. 

It, is well to realize that the stagnant 
layer may not always be the controlling 
factor, so that care must be taken in 
applying this theory to other mass 
transfer problems. 

It appears from theoretical considera- 
tions and experimental verifications that 
the law which governs mass transfer and 
heat transfer can be formulated as 
follows: The action which is obtained 
(the total heat or mass transfer) is equal 
to the driving force (temperature differ- 
ence in case of heat transfer, concen- 
tration difference in the case of mass 
transfer) divided by the resistance (the 
thickness of the stagnant layer). This 
can be stated in abbreviated form as 
follows : 
for mass transfer of gases 



E a = 



B/A 



for mass transfer of liquids 



E a = 



C bm 



B/A 



(1) 



(2) 



In these equations, E a is the total 
number of molecules transferred, which 
is the action obtained, the numerator 
in each case represents the driving 
force, and the denominator, the resistance 
to action formed by the stagnant layer 
of thickness B. 

These basic formulas are found to be 
the governing factors in mass transfer, 
and it can be seen that mass transfer 
can be increased in the following 
manner: 

1. By increasing the driving force. 

2. By reducing the resistance of the 
stagnant layer. 

The resistance of the stagnant layer, 
as a first order approximation, is pro- 



108 



February 1953 Journal of the SMPTE Vol. 60 



portional to the thickness of the stagnant 
layer and, therefore, any methods by 
which the thickness of the stagnant 
layer can be reduced will proportionally 
reduce the resistance of the stagnant 
layer. This proportionality may not 
hold true when the stagnant layer be- 
comes extremely thin, i.e., of molecular 
size, but as a first order approximation 
it will hold true. 

Reduction of Stagnant Layer 

From the foregoing explanation it 
can be seen that the stagnant layer in 
liquids and gases is an extremely im- 
portant factor which determines the 
rate of diffusion. The problem which 
now exists is how to reduce this stagnant 
layer so that the mass transfer can be 
increased. One method of reduction 
of the stagnant layer which has met with 
success is the application of supersonic 
vibrations which so agitate the surround- 
ings as to reduce the thickness. A 
simpler way which has met with greater 
success, however, has been found in 
the application of turbulence. 

Early investigations by experimenters 
interested in the problems of fluid flow 
indicated that if liquid flowed through 
a tube the velocity distribution across 
the tube would be approximately para- 
bolic. The steep sides of the parabola 
represent the stagnant fluid on the sides 
of the wall in which the velocity was so 
low that it could effectively be considered 
as standing still. However, it was soon 
discovered that if conditions in the tube 
were changed and the velocities of the 
liquids increased, a point was reached 
where suddenly the nature of flow 
changed completely. Whereas pre- 
viously the flow was relatively smooth 
and streamlined, the new condition was 
one of extreme turbulence in which a 
considerable agitation of the fluid inside 
the tubes took place, and the velocity 
distribution could no longer be repre- 
sented by a parabola but was more 
rectangular. 

The first condition of flow was arbi- 



trarily called laminar and the second 
condition was called turbulent flow. 
It was later discovered that the Reynolds 
number (Re = DVp/fj.) was the main 
governing factor in the determination 
of whether the flow was turbulent or 
laminar, and it was specifically found 
that a critical Reynolds number exists 
at a value of approximately 2300 above 
which all flow is turbulent and below 
which all flow is laminar. It should 
be noted, however, that conditions can 
be artificially made such that turbulent 
flow can exist at Reynolds numbers 
below 2300, and laminar flow can exist 
at Reynolds numbers above 2300. In 
general, however, the critical Reynolds 
number of 2300 is a reasonable point for 
the determination of the separation 
between the two kinds of flow. 

It is especially important to note 
that the critical Reynolds number (the 
number at which laminar flow is no 
longer possible and turbulent flow 
exists) is 2300 only in circular ducts, 
with fluids flowing axially. For rec- 
tangular ducts correction factors must 
be applied and for circular flow in the 
plane perpendicular to the axis of the 
tube turbulent flow sometimes occurs 
only at extremely high Reynolds num- 
bers (approximately 500,000). Care 
should be taken not to confuse turbulent 
flow with "vigorous agitation." It is 
possible to have vigorous agitation in 
laminar flow, but turbulent flow is a 
specific condition in which the fluid 
provides its own agitation by breaking up 
into a series of vortices. It is only under 
these conditions that the stagnant layer 
is reduced as has been previously dis- 
cussed. 

As a result of a large number of 
measurements in the field of heat 
transfer it was found that the heat trans- 
fer coefficients will be governed by the 
following correlation: 

* 






X 0.023 



Katz and Esthimcr: Turbulent Fluid Processing 



(3) 



109 



where the Prandtl number, in a dimen- 
sionless coefficient, is defined as 



(4) 



Similarly, it was found that the mass 
transfer will be governed by the following 
correlation (which was first pronounced 
by Gilliland): 

1 = i X 0.023 (flt)M(&)M (5) 



where the Schmidt number is 



(6) 



As a result of this formulation, means 
are now indicated for reducing the thick- 
ness of the stagnant layer B. If a 
specific fluid or gas is selected for 
experimentation the following param- 
eters usually remain constant within 
the range of experimentation: 

Viscosity, /* 
Density, p 
Diffusivity, d 
Specific heat, C p 
Conductivity, k. 

As a result the Schmidt number 
which contains the values /i, p, and 5 will 
be constant. The Prandtl number con- 
taining C p) n, and k will be constant. 
Consequently Eq. (3) can then be 
written as follows: 



(7) 



and Eq. (5) can be rewritten as follows: 

(8) 



Usually the less exact notation is pre- 
ferred as follows: 



(9) 



This equation indicates clearly that 
the thickness of the stagnant layer B 
which is a controlling factor in the rate 
of diffusion (and thus the rate of develop- 



ment) is inversely proportional to 
factor Re*-*/D. 

If we have a tube through which 
liquids or gases flow, the mass transfer 
coefficient cannot be indefinitely in- 
creased, as for a given diameter an 
increase in velocity will result in an 
increase in horsepower required to 
drive the gases or fluids through the tube. 
Similarly, a reduction in diameter at 
constant velocity will cause an increase 
in the pressure drop required to force 
the gases or liquids through the tube 
resulting in an increased horsepower 
and equipment size. Consequently, 
there is a practical limitation to the 
choice of the velocity and the diameter. 
Velocities near or above the sonic level 
are usually undesired because of other 
effects accompanying supersonic flow. 
A reasonable compromise, however, 
can usually be found with velocities in 
the vicinity of 50 to 400 miles per hour 
and diameters in the range of 0.1 to 7 
mm. This permits the use of Reynolds 
numbers well in the turbulent region 
in the vicinity of 50,000 to 500,000 and 
Re-*/D factors of considerable magni- 
tude. The horsepower required to drive 
the fluids or gases under those conditions 
is usually quite low, and experiments 
can usually be performed with pumps or 
blowers driven by motors of reasonable 
size (ranging from 1/20 to 2 hp). In 
general, the design of the equipment is 
dictated to a great extent by the avail- 
ability of a pump, and usually the 
design of a turbulent fluid chamber is 
based on a particular pump or blower 
which is available rather than on 
optimum theoretical considerations. 

Application of the Theory 

to Photographic Development 

The foregoing theory indicates that 
improvement can be obtained from the 
application of turbulent fluids to photo- 
graphic development and that this will 
be a function of Re*-*/D. The improve- 
ment, however, will only be significant 
if the following conditions prevail: 



110 



February 1953 Journal of the SMPTE Vol. 60 



1. A substantial concentration gra- 
dient exists through the emulsion. 

2. The diffusion time through the 
emulsion is not abnormally high. 

3. The time of chemical reaction is 
not abnormally long. 

4. The emulsion is relatively thin. 

If any one of these aforementioned 
quantities becomes dominant, so that 
it becomes the main controlling factor 
of the photographic process, the improve- 
ment due to the application of turbulent 
fluids will be considerably reduced. 
Conversely, any reduction in these 
factors will increase the time gain 
obtained by the application of turbulent 
fluids. Consequently, the gain resulting 
from the application of turbulent flow 
will, in general, increase as the speed 
of photographic processing is increased. 
Therefore, greater improvements can 
be expected from the application of 
turbulence if more concentrated, higher- 
temperature developers are used with 
thinner emulsions. 

Description of Equipment 

In order to perform satisfactory 
experiments in which quantitative data 
could be obtained with a reasonable 
degree of consistency, a turbulent proc- 
essing unit had to be constructed which 
would permit changing the desired 
variables while leaving all other param- 
eters constant. The original proc- 
essing unit which had been constructed 
during early experiments (see Fig. 1) 
was not suitable because it did not 
permit switching fluids. Consequently, 
the film samples had to be removed 
from the chamber by hand and be 
immersed in a fixer or stop bath for the 
processing. This hand operation varied 
too much in time from one sample to 
the next and consequently a wide 
spread in the data was obtained. 

It was, therefore, decided to build an 
equipment to conform to the following 
specifications : 

1. The machine will permit the rapid 



introduction and withdrawal of a film 
strip approximately 35 mm wide and 6? 
in. long on which a photographic wedge 
has been exposed. 

2. The machine will permit the 
introduction of different channel diam- 
eters over the film so as to permit the 
study of different Reynolds numbers and 
different diameters. 

3. The machine will permit the 
introduction of several fluids in rapid 
succession. 

4. Timing will be accomplished elec- 
trically and automatically so as to 
eliminate human error. 

5. At least one of the fluids will be 
variable over a wide range in pressure 
and volume. 

6. Provisions will be made for thermo- 
statically controlling the temperature 
of the solutions. 

7. Volume flow, temperature and 
pressure of the fluid will be measured. 

As a result a turbulent processing 
machine was designed as shown in 
Figs. 2 to 8, inclusive. This machine 
consists of three basic units as follows: 

1. Base frame containing motors, 
pumps, drip tank, variable speed drive, 
fluid tanks and cooling system. 

2. Turbulent developing chamber 
with intake, exhaust and by-pass valves. 

3. Control panel containing motor 
and valve timing controls. 

The base frame (see Figs. 2, 5 and 6) 
is of angle-iron construction of dimen- 
sions 60 X 30 X 42 in. Four 1-hp, 
1725-rpm, 220-v, 3-phase, Ideal motors 
are mounted on the lower level of the 
base frame. Each of these motors 
drives an Eastern Industries Model Z, 
flange-mounted, 304 stainless-steel pump 
through belts and pulleys. The pumps 
are mounted several inches above the 
motors, which permits the changing of 
pulleys so as to get different operating 
speeds of the pumps. In addition, a 
Worthington All-Speed drive was in- 
stalled between one of the motors and 
the pump to permit continuous varia- 



Katz and Esthimer: Turbulent Fluid Processing 



111 




Fig. 2. Overall front view of completed high-speed film processor. 



tion of the pump speed from one-eighth 
the motor speed to twice the motor 
speed. A cast-iron protectively coated 
junction box was mounted on the frame 
to allow waterproof electrical connec- 
tions. All wiring was encased in steel 
tubing and completely shielded from 
the splashing of the fluids. A large 
stainless-steel drip pan of 10-gal capacity 
was placed beneath the pumps, above 
the motors, with provisions to drain 
the pan in one corner. 



The chamber and valve frame (see 
Figs. 2-6) is mounted on the top right- 
hand side of the base frame. The 
valves control the introduction of the 
various fluids into the processing cham- 
ber. This processing chamber is made 
of 1-in. lucite, and designed in such a 
way that film can be introduced into 
the chamber easily, while still permitting 
the use of different channel sizes. For 
this purpose a draw slide was built into 
this chamber (see Figs. 3 and 4) in 



112 



February 1953 Journal of the SMPTE Vol. 60 



From 
Film 

Developer Short- Stop 
Pump Pump 



Solenoid-Operated Slide Valves 
From From From 




To Developer To 

Tank Short-Stop 



Solenoid-Operated Slide Valves 



Fig. 3. Turbulent developing chamber 
and slide valves. 



which the film is fastened and onto 
which the turbulent flow channel is 
attached. When the draw slide is 
pushed into the unit it is sealed by 
means of a gasket, firmly connected 
on the outside and clamped by means 
of three screw handles. The fluid 
entrance and exit consist of two 1-in. 
stainless-steel nipples press-fitted into 
two lucite endpieces. Adjacent to the 
endpietes are two plenum chamber 
sections to provide a space in which 
the fluid pressure can be equalized across 
the channel face. From the plenum 
chamber the fluid goes through a narrow- 
ing section until the channel has the 
desired configuration of the channel 
above the film. The film slide contains 



Fig. 4. Cross section through 
turbulent developing channel. 



two clamps in which the film is clamped. 
The film strips are approximately 
in. long by If in. wide. The two 
film clamps are designed in such a 
manner as to cause a minimum of flow 
disturbance. A steel and sponge-rubber 
hinge assembly closes on the outside 
of the chamber to minimize leakage. 

The third component of the equipment 
is the control panel (see Fig. 8). It 
contains all the electrical control com- 
ponents and circuits. The lower four 
buttons control the 1-hp motors which 
operate the pumps. A set of buttons 
across the top of the panel control the 
solenoid valves. The timers are mounted 
below the buttons and a selector switch 
has been mounted so as to permit the 
introduction of the timers into any one 
of the circuits. One tinier controls the 
time during which the fluid is pumped 
through the chamber. At the end of the 
pumping cycle, as selected by the 
selector switch, the fluid valve auto- 
matically closes and an air stream is 
sent through the chamber to flush the 
system for a short period of time. The 



Katz and Esthimer: Turbulent Fluid Processing 



113 







Fig. 5. Turbulent processing unit, rear view of base frame 
with associated plumbing. 




114 



Fig. 6. Base frame of turbulent processing unit. 
February 1953 Journal of the SMPTE Vol. 60 



A!R VALVE LOAOiNG A!R 

CYLINDER SPRING SOLENOID VALVE 




Fig. 7. Detail of upper valve mechanism of turbulent processing unit. 



flushing time period can be set on a 
separate timer in preparation for the 
following fluids. A 30-min timer has 
also been mounted on the panel to permit 
automatic developing for longer periods 
of time. These timers have an accuracy 
of J sec over their entire range. Inas- 
much as it was expected to operate the 
entire equipment in total darkness, 
and in order to permit emergency dis- 
connection in case of failure of the 
equipment, a bar was provided across 
the front of the panel which actuated 
the main contactor, disconnecting all 
valves and motors, when depressed. A 
reset button was mounted behind this 
emergency bar. 

Two pressure gauges were installed 
on the control panel to read the input 
and output pressure of the film chamber. 
A stainless-steel flow meter (Shutte & 
Koerting Co., Size 6 Rotameter) was 
installed to permit an accurate measure- 
ment of volume flow of the fluids going 
through the chamber. 



The valves used in this machine were 
stainless steel, Ij in. in diameter oper- 
ated by compressed air. The air 
cylinders were controlled in pairs from 
a two-way pilot valve in such a way 
that either the by-pass valve or the 
intake valve was open at all times. The 
pilot valve automatically reversed the 
position of the two valves when actuated, 
and a suitable delay was built into the 
pilot valve system so that the by-pass 
valve would not close until the channel 
intake valve was open. The four 
exhaust valves were each controlled by 
their individual electrically operated 
pilot valves. 

All plumbing between the valves and 
the chamber was either stainless steel 
or Carlon.* In addition a water line 
was built into the system controlled by 
a hand-operated stainless-steel valve 

* Carlon Products Co., 10225 Meech 
Ave., Cleveland 15, Ohio. 



Katz and Esthimer: Turbulent Fluid Processing 



115 



Table I. 





Cross-section 
size of channel 


Max. Re 
obtained 


Gpm at 
max. Re 


A p at 
max. Re 


Max. Re - 9 
~JT 


Hp req. 


1 


|X fin. 


Not used 


Not used 


Not used 


Not used 


Not used 


2 


1 X fin. 


85,800 


25.7 


3.75 


1.94 X 10 s 


0.057 


3 


1 X iin. 


90,800 


24.7 


9.0 


2.78 X 105 


0.13 


4 


1 X * in. 


82,600 


20.3 


18.5 


4.64 X 10* 


0.22 


5 


1 X 3^ in. 


51,400 


11.9 


20.5 


6.02 X 105 


0.14 


6 


1 X &in. 


38,700 


8.7 


23.5 


9.40 X 10 6 


0.12 


7 


2 X .015 in. 


11,200 


5.25 


19.5 


6.95 X 10 s 


0.06 



which permitted the flushing of the 
entire system. 

Each fluid tank had a cooling system 
using i%-in. stainless-steel cooling tubes 
through which tap water was forced. 
The flow of the tap water was regulated 
by means of four solenoid-operated 
valves which were controlled from four 
thermostats hung in the individual tanks. 
This system automatically maintained 
the four solutions at between 68 and 
70 F. 

In order to perform proper tests a 
number of different channels were made 
available to be mounted in the draw 
slide over the film. These channels 
were all milled out to a different depth 
and had proper entrance and exit 
sections to reduce the pressure drop. 
The available channels are shown in 
Table I. 

Inasmuch as the total number of 
stainless-steel valves required for the 
operation of four fluids was not imme- 
diately available and time was of the 
essence, the machine as shown in Figs. 
2 to 8 inclusive was constructed for two- 
fluid operation and manual water 
flushing. Most experiments were done 
on the machine as shown and the other 
four valves permitting four-fluid opera- 
tion were not installed until later. 

Description of the Tests 

Definitions. Before describing the tests 
performed it might be well to define 
some terms which will be used hereafter. 

"Stagnant" or "no agitation" de- 



veloping refers to the development of a 
sample of film lying completely at rest 
in a standard developing tray. The 
fluid in the tray is completely at rest. 
The fluid in the tray consists of a sample 
of fluid taken from the developer tank of 
the turbulent processing machine. These 
samples of developers were taken from 
the tank of the turbulent processing 
machine only after the developer fluid 
had been circulated through the machine 
so as to provide thorough mixing. After 
the sample of developer fluid had been 
put into the tray the temperature of the 
fluid in the tray was measured with a 
thermometer. 

"Manual agitation" or "constant 
agitation" refers to a test sample of 
film which is agitated vigorously through 
the developer in a developing tray. The 
developer in the tray was again obtained 
from the tank in a turbulent processing 
machine as before. The temperature 
of the developer in the tray was again 
measured with a thermometer before 
and after tests. 

Resolution Test. A rough resolution 
test was first performed with a number 
of film strips of Super XX Aerographic 
film exposed to a newspaper by means of 
an ordinary camera. A number of these 
strips were placed in a standard develop- 
ing tray where they were developed by 
either "no agitation" developing or 
"constant agitation" developing (for 
definitions of these terms see above). 
A number of similar samples were then 



116 



February 1953 Journal of the SMPTE Vol. 60 



CHAMBER OUTLET 
PRESSURE GAUGE 




Fig. 8. Front view, control panel of turbulent processing unit. 



mounted in the turbulent developing 
machine and developer fluid was run 
over the film at high speed. After 
fixing, washing, and drying, the samples 
were carefully examined and no large- 
scale distortions detectable under a 
microscope were found to exist, so that 
it was possible to continue with the 
remaining tests without any further 
worries of distortion. A sample of this 
test has been reproduced in Fig. 9. 

Turbulent Developing. The turbulent 
developing tests were performed as 
follows : 

A sample of film was taken in complete 



darkness from its storage box and 
mounted in the turbulent developing 
chamber containing a channel pre- 
viously selected. The draw slide was 
pushed in and the gasket seal was 
clamped on. The timer had previously 
been set to the desired turbulent de- 
veloping time and the motor and pumps 
were running. The valves were in the 
by-pass position so that fluid was already 
being pumped around but did not pass 
through the chamber. 

The timer button was then pushed, 
permitting the developing fluid to pass 
over the test sample for the desired 
period of time. After the completion 



Katz and Esthimer: Turbulent Fluid Processing 



117 



irtrowit has what Pubiic 
of the com- han can sper 

54-year-old With 16 Republicans boMng handkerchiefs 
leadership, the House sun & 1 "**- 

Treasury Depart- 1 ^ d wt th * iim ^ * *>' McDermott 
iney**. attorney, C. the budget for Callahan's engt 
ing p 8 rsonnd44,60,000 for 
rmanent employes and $381,000 
i ary help. 
publicans who supported 




s Commr, CaHa- 

jng personnel, clded to P* 



istration numb< 
car, It was beJ 

stolen, 

Minutes aftri 
bandit 
Warren avenue 



Normal Agitation 
D-19 67F 

5 min. 



July 18, 1951 



ns 




us-t :a *ts an u* 

shoremen. 

The bandit 
thin, thp other 
in an autontobi 
co*npiice throii 
yard entrance, 
sive high met 
open aifd no g 
'Itie dock ad 
Naval Shipyar* 

r>* t 



Turbulent Agitation 

D-19 67F 

1-1/2 min. July 18, 1951. 

Fig. 9. Rough resolution test. 



of the timing cycle the fluid was drained 
out of the chamber and air was intro- 
duced to blow the remaining fluid out 
for approximately one second. The 
fixer solution was then introduced into 
the chamber and permitted to fix the 
sample. The time used for fixing was 
always in excess of the maximum time 



required for complete fixing of the 
sample. 

The following control tests were per- 
formed before and after each run. 
Before the beginning of a run of tests 
a number of samples were put in a tray 
for different lengths of time and de- 
veloped with "no agitation." These 



118 



February 1953 Journal of the SMPTE Vol.60 



lamples were then removed and put in 
k tray containing fixing bath and later 
Ivere put in a normal wash tank with 
running water. A number of samples 
Ivere also developed under condition of 
rcontinuous agitation." These samples 
Lvere again fixed in a regular tray and 
kvashed in a regular bath. The same 
brocedure was also followed after the 
Completion of a run of tests and before 
the particular developer used in the 
tests was thrown away. These control 
[tests gave a close check on any small 
variations in the properties of the de- 
veloping fluid. All samples were 
properly marked during the tests so 
that identification was possible after 
the samples had been washed and dried. 

The washing of all samples was done 
in the regular running water tray bath 
for a period of at least one hour and the 
samples were then dried by hanging 
them up on a line. The densities of the 
exposed wedges were read on a densitom- 
eter. 

The film used during the entire tests 
was Super XX Aerographic film which 
was cut up into small samples approxi- 
mately 6 in. long and \\ in. wide. 
These film samples were exposed to a 
photographic wedge in the sensitometer 
at Boston University. The film samples 
were kept in a light-proof box in the 
darkroom and were usually used for the 
tests within a day or two after exposure. 

A number of different channel heights 
were used in the tests with the different 
channels as described previously. In 
addition, for each channel the pump 
speeds, and consequently the flow rate, 
through the chamber were varied con- 
siderably by means of the variable speed 
drive contained in the system. 

The temperature of the operating 
fluids was accurately maintained by 
means of an independent temperature 
control system. This system consisted 
of separate thermostats for each tank, 
operating a solenoid valve which intro- 
duced cooling water through the cooling 
coil in the tank. The temperature 



variations of the fluids as observed were 
approximately J. The temperature 
was maintained at 68 F. 

Turbulent Fixing. Turbulent fixing tests 
were performed in the following manner: 
A strong light was mounted behind the 
turbulent processing chamber and the 
observer placed himself on the other 
side of the chamber, so that he could see 
the sample mounted in the chamber 
with the light shining through it. Turbu- 
lent fixer was then passed through the 
chamber and the observer noted the time 
required for the sample to clear com- 
pletely. As soon as the sample was 
completely cleared the observer would 
notify a second observer who noted the 
exact time elapsed. All the different 
channels were used in the turbulent 
fixing tests under different conditions of 
flow as observed in the flow meter and 
the pressure gages. "No agitation" 
fixing and "constant agitation" fixing 
tests were performed to check on any 
variations in the fixing bath. The 
samples used were not developed to 
eliminate variations introduced by the 
developer. 

Turbulent Washing. For the turbulent 
washing tests a sample was first de- 
veloped in a tray under conditions of 
"no agitation," then fixed in a tray 
under conditions of "no agitation." 
The wet sample was then mounted in 
the turbulent developing chamber and 
turbulent water was passed over the 
sample. It should be noted here that 
the water used for turbulent washing 
was recycled but fresh water was also 
added continuously during the test. 
After a fixed period of time, as deter- 
mined by the electric timer, the water 
flow over the sample was shut off, the 
sample was quickly removed from the 
chamber and mounted in a test tube 
containing silver nitrate. The sample 
was then vigorously shaken until a de- 
posit was formed and the amount of 
deposit was compared with a number 



Katz and Esthimer: Turbulent Fluid Processing 



119 



1.4 



1.3 



1.2 



1.1 



1.0 



0.9 



'0.7 



'0.6 



0.5 



0.4 



0.3 



0.2 



0.1 






7 



7 8 9 10 11 12 13 14 15 16 17 18 19 
Developing Time (Minutes) 



Fig. 10. Plot of average gamma vs. developing time. No agitation; tray develop- 
ing; temp., 68 F; = experimental points; X. = average of experimental points. 



1.4 
1.3 
1.2 
1.1 
1.0 
0.9 
^0.8 
|0.7 
"0.6 
0.5 
0.4 
03 
0.2 
0.1 
























\ 




^** 
































,^ 


**** 






























> 


^ 


































/ 


' 


































/ 




































/ 






































/ 




































I 












































































1 




































1 






































I 






































' 













































































1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 
Developing Time (Minutes) 

Fig. 11. Plot of average gamma vs. developing time. Constant agitation; 

tray developing; temp., 68 F; = experimental points; 

X = average of experimental points. 



120 



February 1953 Journal of the SMPTE VoL 6Q 



4.5 
4.0 
3.5 

5 3 ' 

2.5 
I 
^ 2.0 

1.5 
1.0 
5 






















~ 


3 
















^ 


*^ 


















^ 


*^* 






* 


'=1.0 










X 


^ 


^^ 


^~~ 


' 


~~~ 










/ 


^ 


^ 

x 


















A 


S 























































































































1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12. 



Fig. 12. Plot of time factor 
(turbulent developing time 
compared to "no agitation*' 
developing time) as a func- 
tion of Re-*/D; gamma = 
1.3; X gamma = 1.0. 



-XlO" 



Fig. 13. Plot of time 
factor (turbulent developing 
time compared to "constant 
agitation" developing time) 
as a function of R<P-*/D; 
gamma = 1.3; X gamma 
= 1.0. 



.3.0 



2.5 



2.0 



1.0 




7 = 1.3 



1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 



of standard deposits which had been 
obtained from "no agitation" and 
"continuous agitation" washing tests. 

Results 

Turbulent Developing. The results of 
the turbulent developing experiments are 
shown in Figs. 10 to 15, inclusive. These 
figures contain only the digested data 
obtained from the readings of the various 
samples. Figures 10 and 11 contain 
all the data obtained in the "no agita- 
tion" developing tests and the "constant 



agitation" developing* tests run as control 
tests before and after the turbulent 
developing tests. In order to have a 
basis for comparison, average curves 
were drawn through all these points 
which produced the average "no agita- 
tion" curve and the average "constant 
agitation" curve. These average curves 
were used for comparison whenever the 
control test data in the experiments had 
been spoiled by fogging. In addition the 
average curves were used as a check to 
observe that the daily control tests did 



Katz and Esthimer: Turbulent Fluid Processing 



121 



0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 




0.8 



Fig. 14. Plot of time factor (turbulent developing time compared 
to "no agitation" developing time) as a function of gamma. 



not vary substantially from this average 
test. If a substantial variation was 
noted then the data with this particular 
type of developer were discarded and the 
developer was declared spoiled. Figures 
12 to 15 inclusive contain the time 
factors of turbulent developing over no 
agitation and constant agitation de- 
veloping. The factors are defined as 
follows: 

1. The time factor of turbulent 
developing over constant agitation de- 
veloping is defined as the ratio of the 



developing time between continuous 
agitation developing and turbulent de- 
veloping to reach the same gamma. 

2. The time factor of turbulent 
developing over no agitation developing 
is defined as the ratio of the developing 
time between no agitation developing 
and turbulent developing to reach the 
same gamma. 

Examination of Figs. 12 and 13 
indicates that the time factor appears 
to be a slightly rising function of the 



122 



February 1953 Journal of the SMPTE Vol.60 



4.6 



4.4 



4.2 



4.0 



3.8 



3.6 



3.4 



3.2 



3.0 



X 2.8 



E 2.6 



2.4 



2.2 



2.0 



4.52xl0 5 =.ffe 08 /Z/ 



0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 




3.63xl0 5 =fle 8 /> 

__ I I 
3.46xl0 5 =/?e 8 /Z) 



Fig. 15. Plot of time factor (turbulent developing time compared 
to "constant agitation" developing time) as a function of gamma. 



factor Re**/D. It can be seen that a 
variation in the Re*/D factor between 
2 and 9^ produces a variation in the 
time factor of turbulent agitation over 
no agitation varying from 2.3 to 4.5. 
Similarly, it is apparent that the time 
factor of turbulent developing over 
constant agitation developing varies 
between 1.1 and 2.0 for a gamma of 
1.0 and between 1.25 and 2.2 for a 
gamma of 1.3. 

Attempts to plot the data obtained 
as a function of Reynolds number alone 



resulted in a wide spread in the data. 
This verified the theory which indicated 
that the Reynolds number is not the 
critical variable in the rate of develop- 
ment but that the figure Re-*/D is 
the important factor. (A theoretical 
derivation of the reasons for the use of 
this factor has been given previously.) 
When the Re*/D factor is used as the 
abscissa on points obtained with differ- 
ent channels, they fall into a relatively 
smooth curve. 

It is interesting to note that a variation 



Katz and Esthimer: Turbulent Fluid Processing 



123 



of approximately 5 to 1 in the Re-*/D 
factor produced a variation of approxi- 
mately 2 to 1 in the time factor, either 
when compared to no agitation develop- 
ment or constant agitation development. 
It appears that the time factor is greater 
for higher gamma development than 
for lower gamma development. 

The lower ends of the curves in 
Figs. 12 and 13 are not shown but it 
can be noted that the curves in Fig. 12 
should go through the point 1.0 on the 
ordinate when the abscissa (Re-*/D) 
is 0. This indicates that when the 
fluids are at rest the time factor is 1. 
Similarly, the curves in Fig. 13 should 
go through 0.6 on the ordinate when the 
abscissa (Re^/D] is 0, indicating that 
the ratio between no agitation and 
constant agitation is approximately 0.6. 

In Figs. 14 and 15 are plotted the 
time factors as a function of gamma 
comparing turbulent developing to no 
agitation developing and constant agita- 
tion developing. It can be seen that 
the time factor, in general, increases 
slightly as the gamma increases and 
that this increase is not always constant. 
The dispersions in the curves must be 
attributed to the fact that insufficient 
data were taken, so that a small amount 
of fog or uneven developing would 
cause a considerable uncertainty in the 
data. Figure 15 gives a good indica- 
tion of what type of data can be ex- 
pected. All curves follow a fairly 
reasonable pattern in that the time 
factor increases as a function of Re-*/D. 
The only exception in these curves is 
the second one from the top (marked with 
squares) which was taken with channel 
7. This channel was completely differ- 
ent from all the other channels in that 
it had a width of 2 in. instead of 1 in. 
and its opening depended completely on 
the fluid pressure. It was observed 
that the development across the width 
of the channel was not even under all 
conditions and, therefore, considerable 
variation in data may have taken place. 
All the other curves follow a regular 



pattern except that some of them seem 
to indicate that the time factor vs. 
gamma curve is a linear curve, and the 
upper three indicate that the curve might 
be slightly concave with a minimum 
gamma of approximately 1.0. In- 
sufficient data were taken to determine 
the exact behavior of the curves and 
consequently they were plotted as the 
experimental points indicated. It is 
interesting to note that the curves ob- 
tained with channel 6 (1 X -^ m -) 
at a Reynolds number of 38,700 and 
with 8.7 gpm of developer flowing 
through the chamber showed a time 
factor in excess of 2 over the entire 
range of gamma. The horsepower re- 
quired for this rate of developing was 
approximately ^. 

Turbulent Fixing. The results of the 
turbulent fixing experiments are shown 
in Figs. 16 and 17. It can be seen from 
these figures that the large number of 
results obtained with different channels 
follow a clearly definable pattern when 
plotted as a function of Re*/D. From 
Fig. 16 it can be seen that the clearing 
time for no agitation fixing is 255 sec, 
the constant agitation clearing time is 
158 sec, and the turbulent fixing time 
varies from 100 sec to 68 sec over a 
range of Re-*/D from 0.4 to 8.8. A 
plot of the turbulent fixing time factors 
over no agitation fixing and constant 
agitation is shown in Fig. 17. The time 
factors as shown in Fig. 17 are defined 
as follows: 

1 . The time factor of turbulent fixing 
over no agitation fixing is defined as 
the ratio of the fixing times between 
turbulent fixing and no agitation fixing 
to reach the same clearing point (total 
clearing). 

2. The time factor of turbulent 
fixing over continuous agitation fixing 
is defined as the ratio of the fixing times 
between turbulent fixing and continuous 
agitation fixing to reach total clearing. 

It can be seen from Fig. 17 that the 



124 



February 1953 Journal of the SMPTE Vol. 60 



ZbU 
240 
220 
200 
180 

1 16 

HO 
I' 20 

S 100 



80 
60 
40 


|\ 
















1 






















^-Nc 


> Agitation Clearing Time 
















































































































stant 


\gitati 


jnCle 


aring 


Timp - 




















































\ 






































V 






































J^ 




T= 


. 


"""^ * 


T r- 





s== 


^ . 


^ _ 








I- 


^ 





' 


-^ 


-i- 
























































































































X10' 



Fig. 16. Plot of clearing time as a function of Re- 8 /D X 10~~ 5 (inversely propor- 
tional to stagnant layer thickness); temp., 68 F; fixer, F-6. Note: clearing time for 
dry (not developed) exposed film. 



1 

*- 2 

1 















































































E 


^^ 


- 


- 


i. - 


_ 


>. 


< 


, 


Ti 



rbuler 


rt Fixir 


^ 
ig vs - 



'No A 


> 
gitatio 


- 
n" F' 


*. 
ng 


- - 










/ 






































I 




=aa 


. > 


< 


m=^~ 


<- 


- > 



Tu 



'bulen 


<- 

t Fixin 


-> 
gvs. ' 


<- ' 
Const 


__ ^ 
antAg 


;- 
itatior 


-- 
i" Fixi 


; 

ng 












I 


















































































































D 




L 


2 


3 


4 


56789 



X10" 



Fig. 17. Time factor of turbulent fixing vs. "no agitation" fixing 
and "constant agitation" fixing; temp., 68 F; fixer, F-6. 

Katz and Esthimer: Turbulent Fluid Processing 



125 







t 


















\ 


















\ 


\ 


















\N 


\ 


















S^ 


x 


















"\ 


-5C 


pm 
















26 


Gpm 


























































D 5 10 15 20 



Fig. 18. Time factor (turbulent wash- 
ing time compared to flowing water tank 
washing time) as a function of washing 
time; temp., 68 F; 1 X f channel at 

26 gpm (R<f>*/D = 1.86 X 10 5 ); 1 X 
-^2 channel at 5 gpm (R<P*/D 5.95 X 

105). 



Washing Time (Minutes) 



time factor of turbulent fixing over no 
agitation fixing varies from approxi- 
mately 2.55 to 3.75 (average 3.15). 
Similarly, the time factor of turbulent 
fixing over constant agitation fixing 
varies from 1.78 to 2.35 (average 2.06). 
It is interesting to note that the curve 
shown in Fig. 17 rises rather steeply 
initially, but levels off after an Re*/D 
factor of approximately 10 5 has been 
reached. 

Turbulent Washing. The data obtained 
from the experiments with turbulent 
washing are shown in Fig. 18. The 
time factor of turbulent washing over 
normal tank washing is defined as the 
ratio of the washing times between 
turbulent washing and the washing 
obtained by putting a sample in a tank 
through which fresh water is streaming 
at a moderate rate. The comparison 
is made when the two samples, both of 
which have been developed and fixed 
in the same manner, produce a pre- 
cipitate of silver nitrate of the same 
quantity after washing. Two curves 
are shown in Fig. 18, one for a volume 
flow of 26 gpm with a 1 X f in. channel 
and one for a volume flow of 5 gpm for 



a 1 X ^ in. channel. It can be seen 
that the last curve actually has the 
higher Re*-*/D factor and, therefore, 
rises above the other curve. It can 
be seen that the time factor varies ap- 
proximately between 5 and 2 for the 
first curve and approximately between 
5 and 2\ for the second curve. 

Discussion of Results Obtained 

Examination of Figs. 16 to 18 inclu- 
sive indicates that the speed of photo- 
graphic processing can be adequately 
represented as a function of Re-*/D 
under conditions of turbulent flow. 
Although it is important that the opera- 
tion be performed at high Reynolds 
numbers such that the flow is turbulent, 
it can be noted that high Reynolds 
numbers alone are not sufficient to 
obtain rapid processing. Rapid proc- 
essing will, however, be obtained at 
high values of Re**/D. The relation- 
ship of developing time and the Re-*/D 
factor is in agreement with the theory 
except that the theory predicts a linear 
relationship between processing time and 
Re-*/D. From the data obtained 
far it appears that the time factor is an 
approximate function of (R 



126 



February 1953 Journal of the SMPTE Vol. 60 






An explanation of the variations of the 
results from the theory can be found 
hi the experiments which will be pre- 
sented in a later paper regarding the 
delayed onset of turbulence. A con- 
siderable amount of research is now 
being carried on in the field of heat 
transfer to gain more knowledge re- 
garding the onset of turbulence. It 
may be possible that the experiments 
presently being conducted in the field 
of heat transfer concerning the study of 
turbulence promoters can be directly 
applied to turbulent developing so that 
the time factor will again become a 
linear function of Re os /D. Some indica- 
tions exist that increased rates of develop- 
ment can be obtained by fully developed 
turbulence. 

From the data obtained it appears 
that using Super XX Aerographic film 
and D-19 developer at a temperature of 
68 F the time factor of turbulent develop- 
ing over continuous agitation developing 
will vary between approximately 1.5 
and 2.25 and the time factor of turbulent 
developing over no agitation developing 
varies between approximately 2.2 and 
4.5. It is, therefore, reasonable to 
conclude that a time advantage of 2 
over continuous agitation can be ob- 
tained under field conditions if turbulent 
developing is used. Similarly, a time 
advantage over no agitation of approxi- 
mately 3^ can be obtained under field 
conditions if turbulent developing is 
used. Higher time factors may be 
feasible, especially at higher functions of 
Re"-*/D, but this has not yet been 
experimentally verified. 

Examination of Figs. 16 and 17 
indicates that a time factor of turbulent 
fixing over continuous agitation fixing 
varies between 1.78 and 2.35 with an 
average value of approximately 2. 
Similarly, a time factor of turbulent 
fixing over no agitation fixing of 3.15 can 
be expected under field conditions where 
this time factor actually varied from 2.55 
to 3.75. 

The experiments on turbulent wash- 



ing indicate that using Super XX 
Aerographic film at 68 F the time factor 
of turbulent washing over washing in a 
tank with water flowing continuously 
will be approximately 3j under field 
conditions. The time factor actually 
varied between the values of 2 and 5 in 
the experiments. 

The overall results of the experiments 
performed seem to indicate that a time 
factor of turbulent processing over 
continuous agitation processing of at 
least 2 can be obtained under field 
conditions for the entire photographic 
process using Super XX Aerographic 
film with standard developer (D-19) 
and standard fixer (F-6) at a temperature 
of 68 F. The overall results indicate 
also that a time factor of turbulent 
processing over no agitation processing 
of at least 3j for the entire photographic 
processing can be obtained under field 
conditions using Super XX Aerographic 
film with standard developer (D-19) 
and standard temperature of 68 F. 

The experiments seem to indicate 
conclusively that the rate of processing 
will be a function of Re*-*/D and it has 
also been observed that high Re-*/D 
factors can be successfully obtained with 
extremely narrow channels requiring a 
minimum of horsepower for the pumping 
of fluids. It was observed that the 
highest rates of development were 
usually obtained with a total required 
horsepower of not more than ~ to f 
for a photographic strip 6 in. long and 
approximately 1 in. wide. 

Further Improvements to Be Expected 

The experiments described above 
indicated that a minimum time factor 
of 2 is certainly obtainable using turbu- 
lent flow principles. The fact that 
standard developer (D-l 9) and standard 
fixer (F-6) were used at a standard 
temperature (68 F) indicates that this 
time factor of 2 is independent of other 
improvements which can be obtained 
by other means. Consequently, the 
tune factor of 2 will always be in addi- 



Katz and Esthimer: Turbulent Fluid Processing 



127 



tion to other improvements which can 
be obtained by operation at higher 
temperatures or higher concentrations. 

The theory further indicates that 
the time factor of 2 will considerably 
increase when the entire photographic 
process is speeded up. The speeding 
up of the photographic process, in 
general, might be obtained as follows: 

1. Operation at elevated tempera- 
tures (approximately 85 to 100 F). 

2. Operation with more concentrated 
chemical solutions (both developer and 
fixer). 

3. New chemical compositions such 
as mono-bath or other chemicals which 
may have harmful effects when used 
under conditions of no agitation or 
constant agitation but may be well 
suited under conditions of turbulence. 

4. The application of electric fields 
to increase the ion transfer occurring in 
the photographic process. 

5. The use of developing or fixer 
solutions in which the chemicals making 
up the composite solutions are passed 
over the film in succession without 
having previously been mixed. For 
instance, a developer solution will 
normally consist of various constituents 
such as alkali, developing agents, re- 
ducing agents and retarder. It may 
be possible that a greater increase in 
the rate of developing can be obtained 
by passing through alternate shots of 
various submixtures of these agents of 
different balance at intervals of ap- 
proximately one second. In this way 
it might be feasible to swell the gelatin 
rapidly first, then rapidly introduce the 
developing agent which may have a 
different rate of diffusion than the 
swelling agent, then give a shot of re- 
tarder which again may diffuse through 
the gelatin at a different rate. As a 
result it might be possible that the sum 
total of chemicals deposited in the 
gelatin is identical whereas the total 
rate of transfer is increased because of 
the increased concentration of each of the 
components. 



Acknowledgments 

Most of the work described in this 
paper was done in cooperation with the 
Air Forces Air Materiel Command, 
Wright Field, under contract with the 
Optical Research Laboratory, Boston 
University, with the assistance of D. E. 
Macdonald, R. C. Gunter, Jr., H. 
Gewertz, R. Dussault, H. Howell and 
Capt. F. A. Yochim. 

We are also indebted to personnel of 
the Photographic Laboratory at Wright 
Field, Maj. J. Jackson, Capt. E. Conway, 
J. C. Lewis, W. Levison, J. Phelen, 
I. Weisman and Dr. M. Nagel, who 
assisted in the setting up of the design 
objectives of this program. The assist- 
ance of W. Marvin, J. F. Moore and 
E. P. Cignoni, who helped in the design 
of the equipment and the performance 
of the experiments, is gratefully acknowl- 
edged. We are also indebted to Prof. 
J. Kaye of M.I.T., for his valuable 
consulting services. 

References 

1. L. Katz, "Controlled processing of film 
using turbulent flow phenomena," Phot. 
Eng., 2: No. 3, 89-101, 1951. 

2. Final Engineering Report on Study Program 
Leading to Design Data for a High-Speed 
Film Processing Unit, Raytheon Mfg. 
Company, January 1, 1952. 

3. L. Katz, "Ultrarapid drying of motion 
picture film by means of turbulent air," 
Jour. SMPTE, 56: 264-279, Mar. 1951. 

Discussion 

Paul Ireland (EDL Co.) : You referred to 
the diameter is that the equivalent 
diameter? You don't have an actual 
cylindrical tube. You have a rectangular 
cross section. Is there a mathematical 
equivalent to the diameter? 

Mr. Katz: Yes, we use the equivalent 
diameter D e which is given by the following 
equation : 

2Z):Z) 2 



where DI and D 2 are the sides of the 
rectangle. This equivalent diameter is 



128 



February 1953 Journal of the SMPTE Vol. 60 



also four times the hydraulic radius, 
whichever way you'd like to express it. 

Mr. Ireland: From this equation that 
you give here, since the viscosity and 
density are constants, is the diffusion 
proportional to the velocity - 8 then? 

Mr. Katz: The diffusion is equivalent 
to velocity - 8 divided by the diameter 
to the 0.2 power. You may notice that 
the expression governing the rate of mass 
transfer always has diameter in it. So 
it is not sufficient to have high velocity, 
you must also have a very small diameter. 

Mr. Ireland: Were these experiments 
all done with turbulent fluid flow? Were 
there any in which the Reynolds number 
was less than that necessary to give turbu- 
lent fluid flow? 

Mr. Katz: No, the machine was de- 
signed so as to operate at a Reynolds 
number of at least 10,000, which is well 
in the turbulent region. 1 0,000 to 1 00,000 
is the range in which we operated. 

Charles N. Edwards (Fairchild Camera 
Corp.) : I was especially curious about 



your talk on further research when you 
mentioned using electrical fields to accel- 
erate ionization. Can you give me more 
information in regard to your work with 
this process? 

Mr. Katz: The biggest ion transfer in 
the photographic process is, of course, 
washing and we simply computed that 
the rate of diffusion of the ions through the 
gelatin would be helped if an electric 
field would be applied to them. Then 
once they got to the surface they could be 
carried off by the turbulent stream, so in 
combination of electric field and turbulent 
flow you might get a very large decrease 
in the washing time required and therefore 
you would save a lot of water by doing so. 
No further work has been done on this. 

William H. 0/enhauser, Jr., Consultant, 
New Canaan, Conn.: (In accordance with 
a suggestion by Mr. Offenhauser, the 
author has augmented his original defini- 
tions in order to make the presentation as 
nearly as possible complete within itself.) 



Katz and Esthimer: Turbulent Fluid Processing 



129 



Isotransport Camera 

for 100,000 Frames per Second 

By C. DAVID MILLER and ARTHUR SCHARF 



A detailed description is given of the theory, construction and application of 
a new type of Isotransport camera using the rotating drum principle and 
designed to provide a potential repetition rate up to 100,000 frames/sec. It 
is expected that this camera will be of particular interest to organizations 
whose need for equipment in the higher speed range does not justify the 
investment required to provide a permanent installation. 



M, 



.OST OF THE USEFUL applications of 
high-speed photography involve the 
speed ranges covered by available com- 
mercial equipment, from 2,000 to 1 5,000 
frames/sec. However, a considerable 
need exists for equipment usable within 
the next higher order of speed, from 
10,000 to 100,000 frames/sec. Although 
excellent equipment can be constructed 
for use in this higher speed range, the 
investment cost is necessarily high as 
compared with that of the available 
commercial equipment for use in the 
lower speed range. Because of the 
somewhat less frequent need for equip- 
ment in the higher speed range, an 
individual commercial organization may 
require such equipment throughout an 
interval of only a few months within a 
ten-year period. For such limited use 



Presented on October 9, 1952, at the So- 
ciety's Convention at Washington, D.G., 
by C. David Miller, who read the paper, 
and Arthur Scharf, Battelle Memorial 
Institute, 505 King Ave., Columbus 1, 
Ohio. 



the investment cost is often prohibitive; 
hence, desirable and useful research or 
development projects remain undone. 
As a remedy for this undesirable condi- 
tion, Battelle has developed and con- 
structed a new type of Isotransport 
camera for operation in the higher 
speed range. This camera is now 
available for use in research or develop- 
ment for industrial or Government 
sponsors. Use of the camera is expected 
to range from a few months to several 
years for each sponsor, at a cost to each 
sponsor for the camera itself of only a 
small fraction of the total investment. 

Cameras of several types have been 
developed for operation in the speed 
range from 10,000 to 100,000 frames/sec. 
A judgment of the excellence of these 
cameras will rate them in various orders, 
depending upon the particular desirable 
feature used as a basis of comparison. 
The Isotransport principle, selected as a 
basis for the design of the Battelle 
camera, offers an exceptionally good 
compromise between the following de- 



130 



February 1953 Journal of the SMPTE Vol. 60 



jfeirable, and often mutually conflicting, 
I characteristics: 
I High optical speed, 
((High mechanical speed, 
Good resolution, 
Freedom from distortion, 
IjFreedom from blurring due to relative 
motion between image and film, 

I Adaptability to various types of photog- 

raphy and subject matter, 

II Large number of pictures exposed in one 

sequence, 

i Projectability of photographs as exposed 
without re-registration, 

I Simplicity and ruggedness of construc- 
tion, 
Economy of construction, and 

I Economy in the use of photographic 
film. 

The Isotran principle was first used 
in a camera developed in the labora- 



tories of the National Advisory Com- 
mittee for Aeronautics. 1 - 2 That camera, 
of which three units were constructed 
by NACA, has given highly satis- 
factory service over a period of about 
thirteen years. Many of the results of 
its work have been published. 8 Its 
maximum speed is 40,000 frames/sec. 

In the Isotran camera developed at 
Battelle, the mechanical and optical 
arrangements have been considerably 
modified from those of the NACA 
camera, to permit a potential increase 
of repetition rate to 100,000 frames/sec. 

The unique feature of the Isotransport 
camera is the transport of film and 
photographic images by the same optical- 
mechanical part. By this means identity 
of motion of film and image, both as to 
speed and direction, is permanently and 
rigidly assured. Backlash of gears, 
increasing with wear, is never involved. 



ISOTRANSPORT DESIGN 



General Design Features 

Figure 1 is a photograph of the 
Battelle Isotran camera. The visible 
major components are labeled in the 
photograph. The only moving part 
that is photographically functional is 
a disk, approximately two feet in diam- 
eter, which spins within the rotor 
housing. This disk, weighing about 
50 Ib, is fabricated of forged aluminum 
alloy; it turns at 12,000 rpm for a 
repetition rate of 100,000 frames/sec, 
or at proportionately lower speeds for 
lower repetition rates. The disk carries 
either one or two strips of standard 8mm 
film, approximately six feet in length, 
in closed circular loops on the inner 
surface of ledges at its periphery. 

The spinning disk carries, in addition 
to the film, a series of small mirrors 
disposed about its outer periphery. 
These mirrors operate in pairs, one pair 
for each frame to be exposed, a total of 
1,000 mirrors for the 500 frames that 
can be exposed in one photographic 
sequence. These mirrors, which will be 



explained in detail later, function in 
conjunction with a number of stationary 
parts of the optical system of the camera, 
and they constitute the image-transport 
mechanism. Hence the spinning disk, 
with the attached reflectors, constitutes 
both the image-transport and the film- 
transport mechanism. It is from this 
advantageous condition, allowing no 
conceivable variation in relative speed 
of image and film, that the name 
"Isotransport" is derived. 

The spinning disk is driven by a piston- 
type hydraulic motor which occupies 
part of the space within the hydraulic- 
motor housing. This hydraulic motor is 
geared to the shaft of the spinning disk, 
to provide 3^ turns of the disk for each 
turn of the motor. Fluid is delivered 
at 1000 psi to the hydraulic motor by the 
vane pump, direct-connected to the 
three-phase induction motor. 

The hydraulic drive, as opposed to a 
direct or geared connection of the 
electric motor to the spinning disk, 
represents the most convenient solution 



Miller and Scharf : Isotransport Camera 



131 



FIRST SURFACE 
MIRROR 



OBJECTIVE LENS 



HOUSING OF \- * ^ . 

OPTICAL SYSTEM - V^J^ 

Xl \ 



\ 




ULIC "-HSLRVOIR 



Fig. 1. The Isotransport camera developed at Battelle. 



.SUBJECT 




OBJECTIVE LENS 



FIRST REFOCUSING LENS 
SECOND REFOCUSING LENS_ 




FILM AND FINAL IMAGE 



Fig. 2. Simplified schematic diagram of optical system of Isotran 
camera in perspective. 



to the problem of transmitting mechani- 
cal power into an evacuated space. 
The problem of drive is entirely one of 
acceleration to operating speed within 
a reasonable time, against the forces of 
inertia and windage. After operating 
speed is reached, the driving power is 
shut off and the entire series of pictures 
is exposed immediately. After the 
pictures are exposed, the electric motor 



and the vane pump are stopped, and 
the hydraulic piston motor is allowed 
to operate as a pump. Rapid decelera- 
tion is then achieved by constriction of 
the flow of fluid in the discharge line 
from the hydraulic piston motor. 

Also contained within the hydraulic- 
motor housing is a tachometer genera- 
tor, connected with an indicator on the 
instrument panel to show speed in rpm. 



132 



February 1953 Journal of the SMPTE Vol.60 



The optical system, indicated in 
Fig. 1, comprises five stationary lenses 
and five stationary mirrors, in addition 
to the moving reflectors mounted in the 
periphery of the spinning disk. 

The entire assembly, including the 
30-gal reservoir for the hydraulic system, 
is mounted on a structural steel frame- 
work, which in turn is supported on 
casters. The camera, weighing about 
one ton with fluid, can be pushed about 
readily by hand. It has been moved 
over distances of 100 miles by truck 
without any need for readjustment of 
the internal optical system. 

Optical Design 

Figure 2 illustrates the Isotran 
principle in its simplest form. The main 
difference between this simplest form 
and the form actually used is the omis- 
sion of several stationary mirrors, whose 
only function is to bend the optical path 
for mechanical convenience. 

The essential optical components are: 
(1) an objective lens, (2) the moving 
mirrors providing image transport, (3) 
a first refocusing lens, (4) a second re- 
focusing lens and (5) the moving photo- 
sensitive film. 

The objective lens forms a primary 
image of the subject to be photographed. 
This primary image is formed at the 
location of the mirrors that provide 
image transport. These moving mirrors 
impart a motion to the primary image, 
without rotation, as viewed from the 
direction of the two refocusing lenses 
and the photosensitive film. This trans- 
lational motion imparted to the primary 
image is in a direction at right angles 
to the optical axis of the first refocusing 
lens. The purpose of the two refocusing 
lenses is to gather the light from the 
moving primary image, to refocus this 
light to form a secondary and final 
image on the moving film, and to modify 
the direction and speed of motion of 
the secondary image relative to the 
direction and speed of motion of the 
primary image in such a way that the 



.90-DEGREE REFLECTORS 




90-DEGREE REFLECTORS 



Fig. 3. Means of imparting translational 
motion to reflected ray of light. 

final image on the film will move at the 
same speed and in the same direction 
as the film. 

Figure 3 illustrates the manner in 
which the mirrors in the periphery of 
the spinning disk impart translational 
motion to the primary image formed 
by the objective lens, as viewed from 
the position of the first refocusing lens. 
In this figure, the solid lines show a 
pair of mutually perpendicular mirrors 
in one position, and the dashed lines 
show the same pair of mirrors in a later 
position. A single incident ray of light 
is shown, which may be considered as 
having come from the objective lens. 
Two positions are shown, however, for 
this same ray of light after having been 
reflected by both mirrors of the pair. 
The upper emerging ray is in the posi- 
tion corresponding to the first position 
of the mirror pair, the lower emerging 
ray in the position corresponding to 
the second position of the mirror pair. 
By simple geometry, the second emerging 
ray may be shown to be displaced from 
the position of the first emerging ray 
by a distance twice as great as that 
through which the mirror pair moves in 
passing from the position of the solid 
lines to the position of the dashed lines. 

In Fig. 3, for both positions of the 
emerging ray, the incident ray is re- 
flected first from the lower mirror of 
the pair, then from the upper mirror. 
At a later stage in the motion of the 



Miller and Scharf: Isotransport Camera 



133 



SECOND REFOCUSING LENS 
FIRST REFOCUSING LENS 



A 




PRIMARY IMAGE PLANE 



Fig. 4. Manner of modification of 
image motion by refocusing lenses. 



mirror pair, the order of reflection would 
be reversed, the incident ray first strik- 
ing the upper mirror, then the lower. 
However, no change occurs in the 
manner of movement of the emerging 
ray at the time of reversal of the order 
in which the mirrors function. 

As similar translational movement is 
imparted to each ray of light proceeding 
from the objective lens to the primary 
image, upon reflection from the pair of 
moving mirrors, the primary image as 
represented by the emerging rays must 
appear to move in the same direction 
and at twice the speed of the moving 
mirrors. 

Figure 4 illustrates the manner in 
which the refocusing lenses modify the 
motion of the final image on the photo- 
sensitive film relative to the motion 
imparted to the primary image by the 
moving pair of mirrors. This figure 
represents a condition, approximated 
in the actual camera, in which the speed 
of the photosensitive film is the same as 
the speed of the mirror pairs. For this 
condition, the focal lengths of the re- 
focusing lenses and their positions 
relative to the primary and secondary 
images must be such that the secondary 
image will be exactly half as large in 
any linear dimension as the primary 
image. Both primary and secondary 
image, as shown, include an arrow with 
a cross alongside at each end. As may 
be readily seen, remembering that the 
arrow in the primary image is twice as 
long as in the secondary image, the two 
crosses are twice as far apart in the 




SECOND REFOCUSING LENS 



Fig. 5. Schematic diagram of optical 
system as used in the Isotran camera 
at Battelle. 



primary image as in the secondary. 
Hence, if one of the crosses in the pri- 
mary image is regarded as a displaced 
position of the other, the displacement 
is half as great in the secondary image. 
Hence, the rate of displacement of the 
secondary image is half that of the 
primary image. 

Besides the function of imparting 
apparent motion to the primary image, 
and consequent real motion to the final 
image on the moving film, each pair of 
mutually perpendicular mirrors serves as 
a focal-plane shutter governing the 
exposure on the moving film of the 
particular frame with which that pair 
of mirrors is concerned. As each pair 
of mirrors sweeps across the primary 
image, exposure of the different parts 
of the final image on the film proceeds 
in exactly the same manner as if the 
image of the reflectors themselves formed 
the moving slit of a focal-plane shutter 
on the film. All pairs of mirrors start 
and end the exposures of frames at the 
same absolute locations. The continuous 
motion of the film, therefore, effects the 
necessary displacement of the successive 
frames one from another throughout the 
length of the strip. 

Figure 5 illustrates schematically the 



134 



February 1953 Journal of the SMPTE VoL 60 






OPTICAL AXIS 
-OBJECTIVE LENS 



CYLINDRICAL LENS 



FIRST REFOCUSINC LENS 



TWO FIRST SURFACE MIRRORS 
AT 90 DEGREES (ONE NOT 
SHOWN) 



SECOND REFOCUSINC LENS 
SPINNING DISK 
TACHOMETER GENERATOR 




Fig. 6. Vertical section of camera, through axis of rotation, showing mechanical details. 



Isotran system in the form used in the 
new camera. This arrangement differs 
from that of Fig. 2 by the insertion of 
five stationary first-surface mirrors, for 
the purpose of bending the optical path 
around a nearly closed path, in such a 
manner that the moving pairs of mu- 
tually perpendicular mirrors can be 
carried in the outer periphery of the 
spinning disk, while the photosensitive 
film is carried on the inside of a ledge 
forming an integral part of the same 
spinning disk. 

With this arrangement, the first and 
second refocusing lenses are mounted 
with parallel optical axes. One sta- 
tionary mirror is used in each of three 



positions: between the primary image 
and the first refocusing lens; between 
the two refocusing lenses, on the axis of 
the first refocusing lens; and between 
the second refocusing lens and the film. 
Two stationary mirrors are used in the 
position between the two refocusing 
lenses, on the axis of the second re- 
focusing lens. These two mirrors, mu- 
tually perpendicular, are used, instead 
of a single plane mirror, to reverse the 
direction of movement of the final image 
on the film. Otherwise, the final 
image would move at the same speed 
as the film but in the opposite direction. 
Figure 6 is a somewhat simplified 
longitudinal section through the axis 



Miller and Scharf: Isotransport Camera 



135 



of the spinning disk and through the 
axes of the various optical parts, showing 
the important mechanical details of 
this Isotran camera. In this sketch 
two cylindrical lenses are shown which 
were not included in the diagram of 
Fig. 5. 

The cylindrical lens near the focal 
plane of the primary image causes the 
primary image to conform to the 
cylindrical surface traced by the lines 
of intersection of the pairs of mutually 
perpendicular mirrors in the periphery 
of the spinning disk. Upon the second 
passage of the light beam through this 
cylindrical lens, after reflection from the 
moving mirrors, the light rays are re- 
stored approximately to the condition 
corresponding to a flat primary-image 
plane. 

The cylindrical lens near the moving 
photosensitive film causes the final image 
to conform to the cylindrical shape of the 
moving film. 

The emulsion of the photosensitive 
film, carried on the inner surface of the 
ledge, is very slightly closer to the axis 
of spin of the disk than are the lines of 
intersection of the moving, mutually 
perpendicular, mirrors. Any linear 
dimension of the final image must, 
correspondingly, be slightly less than 
half the same dimension in the primary 
image. Ready adjustment of the final 
image to the exact size required is 
made possible by the condition of paral- 
lelism of the optical axes of the two 
refocusing lenses. The three stationary 
mirrors located in the optical path 
between the two refocusing lenses are 
mounted on a single bracket, which can 
be adjusted in a direction parallel to 
the axes of the refocusing lenses. If the 
first refocusing lens is held stationary 
while the bracket mounting the three 
stationary mirrors is moved outward, 
and if the second refocusing lens is 
moved along its optical axis in such a 
manner as to retain a sharp focus of the 
final image on the film, the size of the 
final image changes. 



The spinning disk is axially symmet- 
rical. A second optical system is now 
being constructed, which with the 
present system will allow photographs 
to be taken simultaneously on film 
strips mounted on the inner surfaces 
of the ledges on both sides of the disk. 
This second optical system will be 
located at a position along the circum- 
ference of the rotating disk 25 away 
from the present system. The second 
optical system will be reversed from 
right to left relative to the first optical 
system shown in Fig. 1 . 

The possibility of taking two series 
of photographs of the same phenomenon, 
with an absolute chronological correla- 
tion, will be of great advantage in the 
study of combustion in an engine 
cylinder, and possibly also for other 
applications. In the study of com- 
bustion, schlieren photographs will be 
taken at the same time as direct-flame 
photographs. The results should answer 
a number of questions relative to the 
meaning of phenomena observed in the 
schlieren photographs but not seen in 
direct-flame photographs. 

Optical-Mechanical Details 

The lenses used in the present optical 
system of the camera are all standard 
products purchased from the Bausch & 
Lomb Optical Go. The objective lens 
is a Tessar, //4.5, of 139-mm focal 
length. The first refocusing lens is a 
Baltar, //2.7, of 152-mm focal length; 
the second refocusing lens is a Baltar, 
//2.3, of 75-mm focal length. These 
lenses were selected partly on a basis 
of high optical speed, but primarily on 
the basis of optimum resolution. 

For the second optical system of the 
camera, now under construction, pri- 
mary emphasis will be laid upon high 
optical speed, with good resolution as a 
secondary consideration. At the same 
time that the present optical system is 
used for schlieren photography, the 
second system will be used for direct- 
flame photography, that is, for photog- 



136 



February 1953 Journal of the SMPTE Vol.60 



.90 -DEGREE REFLECTORS 



I Fig. 7. Portion of periphery of spin- 
ning disk showing form of mutually per- 
pendicular reflectors and manner of 
mounting with notched retaining sectors. 



RETAINING SECTOR 




aphy of flame by the light which the 
flame itself radiates. Because securing 
adequate exposure densities is very 
difficult in direct-flame photography, 
some sacrifice in resolution appears 
justified. 

In the second optical system, the two 
cylindrical lenses will be omitted. Al- 
though the resulting images will not 
conform exactly to the rotational surface 
of the moving optical parts, detracting 
somewhat from the quality of the 
resolution, a saving of approximately 
15% of the light passing the objective 
lens will result. For the second optical 
system, the objective lens will be a 
Baltar, f/2.7, of 152-mm focal length. 
The first refocusing lens will be a Cine- 
Dhor, //2.6, of 6.00-in. focal length. 
The second refocusing lens will be a 
Supercinephor, //2.0, of 2.75-in. focal 
.ength. 

A suitable method for fabrication and 
mounting of the mutually perpendicular 
mirrors in the outer periphery of the 
spinning disk presented a perplexing 
problem. Figure 7 represents a portion 
of the periphery of the spinning disk 
showing the form of the mutually per- 
pendicular reflectors and the manner 
of mounting with notched retaining 
sectors. This figure shows a wide slot 
in the peripheral surface of the disk, 
undercut at each side, in which reflecting 
prisms and prism-retaining sectors are 
mounted. As may be seen from the 
figure, the ends of the reflectors extend 
into the undercuts of the slot in the 
peripheral surface of the disk, and are 



held in position by the aluminum re- 
taining sectors which also fit into the 
undercuts. After assembly, each of the 
aluminum retaining sectors is secured 
in position by three No. 0-80 machine 
screws, which are locked by a drop of 
Glyptal on the head. 

When the disk attains a speed of more 
than a few revolutions per second, 
centrifugal force drives each prism-type 
reflector firmly against its seat in the 
prism-retaining sectors. Each prism- 
type reflector, fabricated of a high-alloy 
steel, is polished and coated for high 
reflectivity on one surface only. The V- 
notches were cut in the prism-retaining 
sectors with an accuracy of 2 or 3 
min of arc. As the reflecting surfaces 
themselves are driven by centrifugal 
force firmly against the notched surfaces 
in the retaining sectors, the 90 angle 
between the two reflecting surfaces of 
a pair is accurate within the same 
tolerance as that maintained for the 
V-notches in the prism-retaining sectors. 

Although the arrangement of mount- 
ing of the 500 pairs of mutually per- 
pendicular reflectors in the periphery 
of the spinning disk permits use of prisms 
having only one polished reflecting sur- 
face, the requirements for uniformity of 
dimensions in these prisms presented a 
rather difficult problem. Eventually, 
a remarkably easy solution to this prob- 
lem was found, by which the prisms 
could be manufactured at a compara- 
tively low cost and with extreme uni- 
formity. 

According to this method, the prisms 



Miller and Scharf: Isotransport Camera 



137 



u 




Fig. 8. Finished block of 
60 reflecting prisms. 

were fabricated in blocks of 60, as 
illustrated in Fig. 8. From plates of an 
air-hardening alloy steel, 44 blocks 
were cut, each approximately 2^ in. 
square and -in. thick. The blocks 
were then divided into 22 pairs and 
next were heat-treated to 50 Rockwell 
G and ground all over. One entire 
surface of each block of a pair was then 
covered, by grinding, with V-grooves. 
The grooved surfaces of the two blocks 
were then cemented together, the grooves 
interlocking. This assembly was then 
placed on a magnetic chuck, and the 
upper plate was removed by grinding 
until nothing was left of it except prisms 
of the desired cross section, but more 
than three times the desired length. 
Finally, four grinding cuts were made 
across the face of the block and the 
imbedded prisms, to cut three prisms of 
the correct length from each long prism. 

After completion of the grinding 
operations, the entire surface of the 
block with the imbedded prisms was 
polished optically. A vacuum-deposited 
layer of aluminum was then applied, 
and, finally, a protective layer of quartz. 
The finished prisms were then pried 
out of the matrix, with some aid by a 
solvent to loosen the adhesive. 

Assembly of the reflectors in the 



periphery of the spinning disk involved.] 
laying 50 of them in place at a time, 
then inserting an aluminum retaining 
sector at each end and securing with I 
screws. 

Tests have shown, for the Isotran. 
ca^nera at Battelle, a resolution in the 
projected motion pictures of 20 lines/mmij 
lengthwise of the film and 15 lines/mm i 
crosswise. The resolution in individual- 
frames viewed as stills is slightly less. 

In the earlier form of the Isotran 
camera, the two mutually perpendicular* 
reflecting surfaces used for image trans- i 
port were two faces of a single glass) 
prism. Reflection was internal, and! 
the light rays travelled a distance of] 
about 0.15 in. through glass. 

A difficulty was always encountered^ 
in obtaining glass prisms of this miniature 
type with a 90 angle between the re- 
flecting surfaces accurate to within i 
15 min. If this angle is not exactly 
90, the Isotran camera produces not 
one image on a film frame but two| 
superposed images. The displacement j 
between the superposed images is pro-' 
portional to the error in the 90 angle, 
and to the distance between the focal; 
plane of the primary image and the line i 
of intersection of the two reflecting! 
faces. Hence, if the focal plane could 
be made to contain the line of inter- 
section of the reflecting faces, the two 
superposed images would be identical 
on the film regardless of any small 
error in the 90 angle. However, as 
the reflected light beam passing to the 
first refocusing lens must be separated 
from the incident light beam proceeding 
from the objective lens, it is only possible 
to form the primary image in a focal 
plane that is pierced by the line of in- 
tersection of the two reflecting faces. 
Hence, the two superposed images on 
the film can be made identical over 
their entire area only if the 90 angle 
between reflecting faces is exact. 

With the arrangement in the new 
Isotran camera, in which the two 
reflecting faces are surfaces of inde- 



138 



February 1953 Journal of the SMPTE Vol. 60 



endent prisms, the 90 angle is fixed 
y centrifugal force at exactly the same 
blue as the angle between the corre- 
konding two surfaces in the notched 
gaining sectors. As these notches are 
Si cut with the same tool, it has proved 
ossible to maintain the 90 angle with 

much greater precision than 15 
lin. The separation between the two 
iperposed images on the film is thereby 
reatly reduced. 

In the older Isotran camera, the 
^fleeting prisms are carried on the 
mer surface of a ledge, alongside the 
hotosensitive film. Light from the 
bjective lens is reflected from the 
risms directly across the rotating disk, 
irough the two refocusing lenses, to 
ic photosensitive film surface at the 
pposite side. With this arrangement, 

is necessary that the prisms be set 
n the inner surface of the ledge with 
lie lines of intersection of the mutually 
erpendicular reflecting surfaces at a 
onsiderable angle to the axis of spin of 
le disk. The motion of the prisms 
irough the light beam from the objec- 
ve lens includes a component of rota- 
on about the optical axis of that lens, 
""his rotational component of the prism 
aotion imparts a rotation to the final 
mage on the film. This aberration 
as been eliminated in the new Isotran 
amera by the mounting of the re- 
ectors on the outer periphery of the 
isk, with the lines of intersection of the 
nutually perpendicular reflecting sur- 
aces parallel to the axis of spin of the 
isk. 

In the older Isotran design, with 
noving prisms of triangular cross section, 
11 rays of light pass through one glass-air 
urface of the prism twice in order to 
each and to return from the reflecting 
urfaces. Upon each passage through 
lis surface, part of the light is reflected 
nd reaches the film to form a stationary 
.nal image, uncompensated for film 
novement. Consequently, highlights 
re sometimes recorded as longitudinal 
treaks across the film. This condition 



has been alleviated in the new Isotran 
design, in which no transmission through 
glass is involved in the image-transport 
mechanism at the outer periphery of 
the spinning disk. 

The cylindrical lens located near the 
plane of the primary image of the new 
Isotran camera reflects part of the light 
beam, just as do the surfaces of the 
moving prisms in the older design. 
However, this reflected light is not 
brought to such an undesirably sharp 
focus on the film, and for the most part 
it falls alongside the film rather than on 
the film. 

Mechanical Design 

The camera is 80 in. long, 32 in. 
wide, and 35 in. high. The axis of the 
objective lens of the present optical 
system is in the same vertical plane as 
the axis of spin of the disk, and is inclined 
12j to the vertical. With an external 
mirror above the objective lens, the 
horizontal line of view is 38 in. above 
the floor. External connections are 
made to a 220-v, 3-phase, power source 
for driving the motor. Also, an elec- 
trical circuit is provided for tripping 
the shutter above the objective lens, 
which limits the overall time of operation 
to a single rotation of the disk. A 3-hp 
Stocks vacuum pump has proved very 
satisfactory to evacuate the chamber 
to minimize windage losses. It provides 
a 25-in. vacuum within the housings for 
the rotating disk and the hydraulic 
piston motor within a few seconds. 

The spinning disk, machined from a 
forging of aluminum alloy, was designed 
for service up to 12,000 rpm, correspond- 
ing to 100,000 frames/sec. The rough 
forged blank for this disk was supplied 
by the Aluminum Company of America. 
The material, 14S-T61, has a certified 
yield strength of 61,500 psi, with 12% 
elongation. After forging, the blank 
received a primary heat treatment, 
rough machining, a final heat treatment, 
and finish machining. The forging was 
checked by ultrasonics and found free 



Miller and Scharf: Isotransport Camera 



139 



from defects. Two bands were left on 
the sides of the disk for removal of ma- 
terial to correct unbalance. These 
bands were not utilized, however, 
because balancing equipment sensitive 
to an unbalance of 0.1 oz-in. showed no 
sign of unbalance. 

The calculation of disk stresses was in 
accordance with the method outlined 
by Timoshenko. Computations started 
with the centrifugal force exerted by 
the prisms, and worked back toward 
the axis in six concentric bands. These 
calculations showed that the tangential 
stress at the bore, which was the maxi- 
mum stress, was 25,000 psi for a spin 
rate of 12,000 rpm. 

Two critical speeds were determined 
mathematically, at which forward or 
retrograde precessions should have rates 
identical with the rate of spin of the 
disk. The results indicated a critical 
speed for forward precession of 41,200 
rpm, and a critical speed for retrograde 
precession of 4165 rpm. Although the 
critical speed of 4165 rpm for retrograde 
precession is within the operating range 
of the camera, no difficulty has been 
encountered upon acceleration of the 
disk through this speed. No difficulty, 
in fact, was expected. 

The photosensitive film is held in 
proper position, on each side of the 
spinning disk, by a series of 50 aluminum 
pegs, one for each tenth perforation of 
the film. An overlap of several frames 
is allowed when the film is cut to proper 
length in the darkroom. The strip of 
film is brought to the camera coiled in 
a light-tight container, and is placed in 
the camera through handholes in the 
housing few the spinning disk. The 
disk is readily turned by hand as the 
film strip is fed through a handhole and 
pressed onto the pegs one at a time. 
The direction of loading is the same as 
that of operation, so that the trailing 
end of the film overlaps the leading end. 
Centrifugal force ensures that the emul- 
sion is in the focal plane during exposure. 

The disk is mounted on the over- 



hanging end of a 2-in. shaft. It i: 
pressed onto a taper, is secured by twq 
nuts, and is driven by a face-type key 
The shaft rotates in a pair of Class "/ 
precision ball bearings. The preload 
in these matched bearings is produced 
by clamping with inner and outer 
spacers of identical length. Service is 
so intermittent that no heating problem 
is involved. Adequate lubrication is 
provided by three drops of high-speed 
bearing oil inserted into the vacuum- 
tight oil tube each morning. Labyrinth 
seals prevent passage of oil into the 
housing of the spinning disk. 

Ground helical gears are used be- 
tween the hydraulic piston motor and 
the shaft of the spinning disk, and be- 
tween that shaft and that of the tachom- 
eter generator. 

The housing for the hydraulic piston 
motor and the tachometer generator 
was fabricated of Meehanite castings. 
It is vacuum-tight and is pierced by the 
hydraulic lines, electrical leads for the 
tachometer, an access plug for lubricat-l 
ing the tachometer, and a tube for I 
insertion of gear grease. The housing j 
for the spuming disk and its cover were) 
cast of an alloy steel with a yield strength 
of 100,000 psi. Alignment of the! 
bearing housing and of the optical 
mounting pads at the periphery of this 
housing is ensured by five deep ribs. 
The peripheral surface of the housing 
is 1 in. thick, to provide some protection 
against possible failure of the disk. All 
of the optical components excepting 
those on the spinning disk are mounted 
in the housing or upon brackets attached 
to the housing. The cover of the] 
housing of the spinning disk will serve 
as the principal support for the second 
optical system. 

Meehanite castings form the brackets 
for mounting of the mirrors in the 
optical systems. The top bridge sup- 
ports the focusing mount for the objective 
lens, the external mirror for horizontal 
viewing, and, on its underside, the 
mirror located in the optical path from^ 



140 



February 1953 Journal of the SMPTE Vol. 60 



he moving pairs of reflectors on the 
pinning disk to the first refocusing lens. 
Phe side bridge carries the three mirrors 
ocated in the optical path between the 
wo refocusing lenses. The three mirrors 
upported by the side bridge can be 
noved in a direction parallel to the axes 
f the two refocusing lenses by adding 
r removing spacer blocks between the 
iridge and the housing. Side plates 
orm light-tight covers for the optical 
ystem. 

The lower two mirrors mounted on 
he side bridge must be adjustable to a 
angle with great accuracy. The 
nounting of only one of these mirrors 
s attached directly to the side bridge, 
he other mounting is hinged to the 
irst, and is positioned by a differential 
crew, providing an exceedingly fine 
djustment. A simple optical check, 
vithout the need of any optical instru- 
nents but the human eye, permits the 
ine adjustment needed. 

The mounting for the second re- 
ocusing lens is somewhat complicated 
y the need of incorporating the final 
nirror of the system, and the cylindrical 
ens near the film. 

As the angular velocity of the spinning 
isk is very constant during the period 
>f one rotation during which photographs 
re exposed, no provision is needed for 
stablishment of repetition rate on the 
hotosensitive film itself. A reading of 
he tachometer, recorded at the time 
he photographs are exposed, is satis- 
actory. However, a special spark plug, 
lot indicated in the figures, has been built 
nto the camera housing for establishment 
f chronological identity between one or 
nore frames of the high-speed pictures 
ind corresponding datum points in 
>ther records, such as a photograph of a 
:athode-ray trace. 

The design conditions imposed severe 
equirements upon the drive for the 
pinning disk, involving a top speed of 



12,000 rpm, a reasonably short accelera- 
tion period for the 50-lb disk, an effective 
means of braking, and the need for 
smooth acceleration and deceleration 
throughout. An estimate of the power 
required to drive the disk in air yielded 
a prohibitive value of 40 hp. Operation 
of the disk with at least a partial vacuum 
therefore appeared necessary. No 
encouragement was received from the 
manufacturers of seals for the trans- 
mission of power through the rotating 
shaft from atmosphere into the evacuated 
chamber. Although the use of a high- 
frequency synchronous motor within 
the evacuated space, cooled by circulat- 
ing water, showed promise, this solution 
was considered to be somewhat too 
developmental. 

Finally a hydraulic system was chosen, 
to be supplied by Vickers, Inc. This 
system is standard, with the exception 
of the use of a needle valve in the dis- 
charge line of the hydraulic piston motor 
as a means of braking. The 10-hp 
electric motor drives a constant-de- 
livery vane pump. Both motor and 
pump are mounted on a 30-gal reservoir. 
A pressure-relief valve in the line from 
the vane pump to piston motor is set at 
1,000 psi. The flow-control valve pro- 
vides a satisfactory flow rate of fluid, 
regardless of load. The hydraulic piston 
pump, 5 in. in diameter, is rated at 5 
hp at 3600 rpm. 

In operation, the electric motor is 
started and the flow-control valve is 
set for a flow rate corresponding to the 
desired spin velocity of the disk. The 
pressure gauge in the line between vane 
pump and piston motor verifies the 
establishment of driving pressure and 
the cutoff of driving pressure when 
operating speed is reached. The camera 
disk accelerates with this arrangement 
to 6,000 rpm in 3 min., with a 25-in. 
vacuum in the chamber. Deceleration 
time from 6,000 rpm is also about 3 min. 



Miller and Scharf: Isotransport Camera 



141 




Fig. 9. Schlieren photographs of combustion of natural gas in cylinder of Cooper 
Bessemer GMV engine, exposed at 20,000 frames/sec. Order of exposure is from : 
to 20, through Row A, then through Row B, and so on. The igniting spark occurrec 
at unrecorded time at the position shown by S in frame A-3. First visible flame appear 
as dark area in vicinity of ignition spark at about frame C-4, and is clearly visible a 
F in frame E-3. Flame traverses chamber and disappears throughout remaininj 
frames of series. 



142 



February 1953 Journal of the SMPTE VoL 60 



RESEARCH APPLICATIONS 



The new Isotran camera has been 
sed at rates up to 50,000 frames/sec in 

study of scavenging, turbulence, fuel 
njection, combustion, and exhaust blow- 
lown in a cylinder of a GMV engine 
nanufactured by The Cooper-Bessemer 
Corporation. This spark- ignited engine 
s of the two-stroke-cycle type, and uses 
atural gas as a fuel. The bore and 
troke are each 14 in. The engine, 
nost often built in 10-cylinder V ar- 
angement of approximately 1100 hp, 
s in widespread use for the pumping of 
atural gas through distribution pipe- 
ines. All of the processes enumerated 
lave been made visible by the schlieren 
nethod in the high-speed photographs, 
lie photographs have given Cooper- 
Jessemer valuable information con- 
erning these phenomena, and have in 
eneral verified conclusions previously 
cached on other bases and used as 
esign data. The high-speed photo- 
raphs taken in this study over a period 
f two years are the first known of 
jhenomena occurring within engine 
ylinders of comparable size. 

Figure 9 is a reproduction of a high- 
peed motion picture of combustion in 
ie GMV engine, exposed by the 
chlieren method at 20,000 frames/sec, 
'hese photographs are reproduced here, 
n preference to those taken at 50,000 
rames/sec, because the speed of the 
ihenomena observed does not justify 
le higher repetition rate. 

For the photographs of Fig. 9, the 
)osition of the igniting spark was placed 
vithin the field of view, so that ignition 
ag could be determined. The position 
f the igniting spark, at the apparent 
ntersection of the wire electrodes, is 
esignated by the letter S in frame A-3. 

The order of exposure of the frames of 
ig. 9 is from left to right through 
low A, then from left to right through 
low B, and so on. The three round 
lack spots appearing in most frames are 
ic heads of three cap screws holding a 



mirror to the top of the piston. The 
diameter of this mirror, located near the 
edge of the 14-in. piston, is 2 in. A 
glass window of the same diameter, 
mounted in the cylinder head, permits 
the schlieren view into the combustion 
chamber. 

Flame first becomes visible as a small 
dark cloud in the vicinity of the igniting 
spark at about frame G-4. Throughout 
the remaining frames of Row C, all 
frames of Rows D to G, and the first 
five or six frames of Row H, the flame 
travels across the entire visible part of 
the combustion chamber. During the 
remaining frames of Row H and the 
early frames of Row I the flame burns 
itself out in all visible parts of the 
chamber. The mottled appearance of 
the frames in Rows A and B, and in 
Rows J and K, is due to stratification 
of scavenging air, residual gas from the 
previous combustion cycle, and injected 
gaseous fuel. In the projected motion 
pictures, movements of these strati- 
fications are plainly visible. 

Besides its use in the study of phe- 
nomena in the cylinder of the Cooper- 
Bessemer GMV engine, the new Isotran 
camera is being used in a study of the 
fundamental physical nature of knock 
in a spark-ignited piston engine burning 
various types of liquid hydrocarbon. 
This project is sponsored cooperatively 
through the Coordinating Research 
Council, Inc., with financial support 
by the automotive, aircraft and oil 
industries, and by the Bureau of Aero- 
nautics, Department of the Navy. 
Photographs taken on this project are 
not yet available for publication. 

The new Isotran camera is available 
for additional research projects, which 
might be sponsored by industry or 
Government, several of which may run 
concurrently. Most combustion phe- 
nomena, exclusive of explosions, fall within 
a speed range for which this camera is 
best suited. Most large-scale move- 



Miller and Scharf: Isotransport Camera 



143 



ments of mechanical objects can be 
studied with lower repetition rates. 
However, the new Isotran camera is 
expected to have many applications in 
the study of movements of miniature 
mechanical parts. An inverse relation 
only now becoming recognized is the 
one existing between the size of a 
physical object and the repetition rate 
needed to clarify its movements at a 
given velocity. Even when moderate 
linear speeds are involved, repetition 
rates are needed in the range for which 
the new Isotran camera was designed 
in the study of phenomena such as the 
formation of chips by a cutting tool, the 
cutting action of an abrasive particle, 
the motion through air of droplets of 
liquid, the action of the breach mech- 
anism of a small gun, the operation 
of miniature high-speed gears and ball 
bearings, the spinning or drawing of 
fine artificial fibers, and the transference 
of ink or other material from the surface 
of a high-speed roller. The new Isotran 
camera offers the possibility of studying 
these largely uninvestigated phenomena, 
and offers for this purpose resolution 
adequate for most needs. The camera, 
however, is not intended to be competi- 
tive with those commercially available 
for use in the speed range now covered 
by them. 



References 

1. C. D. Miller, "40,000 frames per 
second," PSA Jour., 74: 669-674, Nov. 
1949. 

2. C. D. Miller, "The NACA high-speed 
motion picture -camera-optical com- 
pensation at 40,000 photographs per 
second," National Advisory Committee 
for Aeronautics, Report No. 856, 1946. 

3. C. D. Miller, "Roles of detonation 
waves and autoignition in spark-ignition 
engine knock as shown by photographs 
taken at 40,000 and 200,000 frames per 



second," SAE Quarterly Trans., 7: 98- 
143, Jan. 1947. 

Discussion 

Carlos Elmer (Naval Ordnance Test Station j\ 
China Lake, Calif., and Chairman of the- 
Session): Can these photographs be printed* 
iri 4 motion picture form? 

Mr. Miller: The photographs are ready 
for projection as a motion picture as theyj 
come from the developer after removal 
from the camera, and they project verjH 
steadily on the screen. 

Anon: How is the film taken up fromi 
the drum onto the spool? 

Mr. Miller: The film is placed in one 
continuous strip around the inside of thd 
ledge on the side of the drum. We take 
photographs only for the time required 
for one complete turn of the drum. The 
drum is brought up to speed; then a 
shutter at the objective lens is opened. 
The shutter stays open for approximately 
the time required for one complete turn 
of the drum. The film is placed in the 
camera by hand, through the handhole, 
and is held in accurate position by pegs 
in the ledge of the drum, one peg for every 
ten perforations of the film. 

Earl A. Quinn (Eastman Kodak Co., Roch- 
ester, N.Y.): It appeared to me from 
your drawings that the film was on the 
opposite side from the prisms. Is that 
true? I understood you to say it was on 
the same side. 

Mr. Miller: Yes and no. In the earlier 
form of the camera, in the laboratories of 
the National Advisory Committee for 
Aeronautics, the prisms and the film were 
both on the inside of the drum, and the 
light went across the drum in passing from 
the prisms to the film. However, in the 
form of the camera I have described here, 
the film is on the inside of the drum and 
the prisms are on the outside. Instead 
of passing across from one side of the drum 
to the other, the light follows a circuitous 
path, from the prisms on the outside of the 
drum to the film on the inside, but with 
the same angular position for prisms and 
film as measured around the periphery 
of the drum. 



144 



Jebruary 1953 Journal of the SMPTE VoL 60 



Photographic Instrumentation in 
the Study of Explosive Reactions 



By MORTON SULTANOFF 



A brief description of explosive reactions is presented to acquaint the reader 
with the problems which arise in the study of explosions by photographic 
techniques. Optical and electronic equipment, designed to produce exposure 
times approaching 10" 9 sec, is discussed from the standpoint of light-gathering 
power, exposure time and cost. Three basic types of cameras (streak, single 
short-duration exposure, and very-high-speed motion picture) are described, 
and examples of typical photographs are presented. 



JL HE APPLICATION of photography to 
the study of explosive reactions is a 
natural consequence of the high lumi- 
nosity which accompanies these reac- 
tions. Three problems of paramount 
importance, with which most other 
photographic studies are not concerned, 
dictate the techniques, equipment, and 
materials employed in the photographic 
studies of explosions. 

First, protection from the blast and 
flying fragments and debris must be 
afforded for both the equipment and 
personnel. Great distances can be used 
to separate the explosion safely from 
the recording equipment. However, all 
the work described in this paper was 
accomplished in a blast chamber built 
directly into an optical laboratory, and 



Presented on October 8, 1952, at the 
Society's Convention at Washington, D.G., 
by Morton Sultanoff, Ballistic Research 
Laboratories, Terminal Ballistics Lab., 
Aberdeen Proving Ground, Md. 



so disposed as to afford adequate pro- 
tection from explosive charges as large 
as 8 Ib at distances as close as 3 ft. 

The very high speeds of the events 
which accompany explosive reactions 
make it essential that the exposure times 
of any photographic recordings do not 
exceed 1/1,000,000 of a second, a unit 
of time which is generally called a 
microsecond. To date, the problem of 
short exposure times has been solved 
only by making necessary compromises 
as described in later parts of this paper. 

Even after the first two problems have 
been solved and successful photographs 
of explosions have been obtained, the 
researcher in this field is faced with the 
task of interpreting the photograph. 
Although much has been learned about 
the identity of the luminosity by past 
investigators, 1 there is still uncertainty 
as to the source of radiation in the 
photographically recorded spectrum in 
many phases of the explosion. 



February 1953 Journal of the SMPTE VoL 60 



145 




1 

- 



g 

2 u 

II 



.s 




o O 



Rankin pinhole 
Bowen multilen 
A.E.G. "Bowen 
O'Brien-Milae i 
dissector 
Sultanoff grid fr 
ing** 
Image converte 









o o o 

O -ri O 



fr 

l b 

1 1 

^ 5 fi 

b, 2 w> 
B O 






itley, 
ror 



and W 
ing mi 
nverter 



.J 

3 to 

^6 

sr*: S3 

IHIJ! 







<vf 

? 

-!^ f 


"O 

E 


13 

J3 
. to 
^ fc^ 


tro-optical 
purposes," 


h-T O 

o;? c 

5|s 

s:** 




QS 


. Rankin, 


per-second 


en* 

! 




o 


NAVORD 


I 
\ 


1 


















. 












t> ^ 




146 



February 1953 Journal of the SMPTE Vol. 60 



In the interest of brevity this paper 
is restricted to those phases of the explo- 
sion which involve the detonation of an 
uncased explosive charge and the shock 
which is observed in the immediate 
vicinity of that charge. The equipment 
and techniques discussed are those with 
which the author is familiar and those 
which, in his opinion, are applicable 
to the studies covered in this paper. 
Because of the requirements of military 
security there are many omissions of 
material and credits to other investi- 
gators engaged in similar work. 

Explosives and Explosive Reactions 

An explosive substance is one which, 
when subjected to the proper initiating 
conditions, undergoes a rapid chemical 
transformation to form more stable 
substances which are mostly gaseous 
and which have a combined volume 
(under normal atmospheric conditions) 
tremendously larger than that of the 
original substance. The chemical re- 
action is usually accompanied with 
temperatures in the range of 3500 K 
and pressures of several million pounds 
per square inch. The progress of the 
icaction or detonation front through 
the explosive is regular and well ordered. 
The detonation front moves with a 
speed of about 25,000 fps (feet per 
second), with the exact speed depending 
on the composition and density of the 
explosive. 

Large volumes of gas are liberated 
by the detonating charge in such a 
short interval of time that a rapidly 
moving surface of discontinuity, across 
which there is an abrupt change of 
pressure and temperature, is formed in 
the air which surrounds an uncased 
charge. The velocity of this "shock 
front" is initially even higher than the 
detonation rate in the explosive (about 
30 times the speed of sound in air), but 
it decays initially very rapidly. As the 
shock velocity approaches the velocity 
of sound the decay rate becomes very 
gradual. It is this shock which inflicts 



the blast damage common to explosions. 
A special application of explosives 
based on the "Munroe Effect" (shaped 
charge) which involves the "focusing" 
of the explosive forces, and which is 
especially effective in the penetration 
of tank armor, has been described 
recently in many unclassified papers. 2 
Since several photographic studies dis- 
cussed in this paper involve the shaped 
charge, it should be noted that when a 
cylindrical stick of explosive with a 
conical metal liner in one end is initiated 
at the opposite end, the detonation 
wave which passes over the metal cone 
causes it to collapse rapidly, and a 
"jet" of small metallic particles streams 
out at very high velocity. This jet 
can penetrate to amazing depths in 
common metals and, employing this 
effect, relatively small explosive charges 
have been used to produce holes through 
the walls of tanks. 

Photographic Instrumentation 

Comparison of Basic Types of Instruments 
for Photographing Explosions. Streak re- 
cording is the most common form of 
photography employed in the study of 
explosive reactions. This type of record- 
ing produces a plot of distance vs. time 
on a single strip of film by the relative 
motion of a narrow transverse slit along 
the film. Rotating mirrors and rotating 
drums have been used to produce the 
motion of the slit with respect to the 
film, and image speeds as high as 10,000 
fps, which produce exposure times as 
short as 10~ 8 sec are readily obtainable 
with relatively simple equipment which 
can be fabricated at a reasonable cost in 
the average research laboratory. This 
type of recording has a serious short- 
coming in that only the action which 
takes place in view of the narrow slit 
is recorded. The streak camera is blind 
to the events which occur out of the 
plane of vision of the slit. 

Single exposures of explosive events 
have been obtained with ordinary pres- 
sor professional-type sheet-film cameras 



Morton Sultanoff: Study of Explosives 



147 



Explosive- . 



Ul 

Initiated end 




SIDE VIEW 



Lens 



Rotating mirror 



Fig. 1. Schematic diagram of the formation of the streak image 
in the rotating-mirror camera. 



by replacing or supplementing the 
standard mechanical shutters by elec- 
tronic shutters which have no moving 
parts. The equipment most widely 
used to obtain submicrosecond single 
exposures involves the interference of 
polarized light. This type of shutter 
is generally expensive, and competent 
electronics personnel are required to 
design, build and operate it. The major 
shortcoming of this equipment is that 
only a single exposure can be obtained 
with each shutter. To obtain data in 
the study of explosives, it is therefore 
necessary to have a whole battery of 
cameras or to fire large quantities of 
like charges and get a single exposure 
from each, timed at a different instant 
after firing. 

The most desirable, but by far the 
most complex and costly of the cameras 
employed in the study of explosives are 
motion picture cameras which can be 
used at rates from 100,000 to 100,000,000 
frames/sec with the exposure times of 
the fastest cameras approaching 10" 9 
sec. Cameras operating in this speed 
range are usually designed for specific 



performance characteristics at the 
expense of other desirable, but less 
essential, characteristics. 

Table I is a summary of the equipment 
discussed above and includes the specific 
cameras now in use at the Terminal 
Ballistic Laboratory, described in the 
following paragraphs. 



Streak-Camera (Bowen RC-3 Rotating 
Mirror}. The earliest type of camera 
used successfully in the study of high- 
speed transient phenomena, 3 and perhaps 
the most widely used today in the study 
of detonation and shock rates, is the 
streak-camera. The photographs ob- 
tained with this camera are records of 
distance as a function of time, and 
therefore differ from standard photo- 
graphs, which are essentially records of 
space taken at a fixed time. 

The essential feature of the streak- 
camera is a method of producing rela- 
tive motion of a slit along the film plane. 
Rotating-drums and rotating- mirrors 
have been used to produce the relative 
motion of the slit along the film, and the 
rotating-mirror has proved to be the 






148 



February 1953 Journal of the SMPTE VoL 60 




a. Streak rote recording. 




Cylinder 





'I 
1 


- 




1 

1 -* 


Hollo 




> ^ ^ 


cylin 



b. End view. 



c. Profile recording. 



Fig. 2. Rotating-mirror camera records: a. rate record of shock to show chang- 
ing velocity; b. end-view record to show curved front by nonsimultaneous 
emergence; and c. profile record to show curved front. 



Morton Sultanoff: Study of Explosives 



149 



?2 




150 



February 1953 Journal of the SMPTE Vol. 60 






superior of these two methods. The 
discussion which follows describes the 
formation and motion of the image in 
the Bowen RC-3 rotating-mirror streak- 
camera. 4 

In all streak-cameras the event (or an 
optical image of the event) takes place 
at the slit. In Fig. 1 an image of the 
slit is formed on the film plane by the 
lens after reflection from the rotating 
mirror. A cylindrical stick of explosive, 
located on the slit and initiated at the 
end as shown in Fig. 1, will light the slit 
at the detonation front I , and an image 
of the lighted portion will be formed 
by the ray RO on the film plane at IV 
At some later time the detonation front 
will have moved along the slit and 
reached a position in the charge indicated 
by Ii. In this period of time the rotating 
mirror will have moved through an 
angle a, and by a simple geometric 
construction it can be seen that the 
position of the lighted portion of the 
image of the slit will fall at I'i. By 
similar construction a continuous series 
of points can be located along the film 
plane, and the streak "S" will be 
formed. Since the axis of the motion 
along the film represents time and the 
motion along the slit represents distance, 
the slope of "S" is velocity when the 
proper scale-distance and scale-time are 
employed in the measurement of that 
slope. By the same type of geometrical 
construction it can be shown that if 
velocity varies along the slit the streak 
will be curved (Fig. 2a). These velocity 
recordings are commonly referred to as 
"rate" records. 

If the axis of a stick of explosive is 
perpendicular to the slit, and the stick 
is initiated at one end, the "profile" of 
the front will be recorded as shown for 
the curved front in Fig. 2c. End-view 
emergence records are also obtainable 
with the rotating-mirror streak-camera 
with the slit oriented as shown in Fig. 
2b. The choice of which of these two 
types of recordings is best suited to a 



particular "profile" study depends on 
the waveshape and data required. 

This simplified explanation of the 
operation of the Bowen RC-3 rotating- 
mirror streak-camera does not include 
a discussion of such details as the actual 
shape of the focal plane which is gener- 
ated by reflection from the octagonal 
mirror which is not rotating around its 
face. That analysis has previously been 
published. 4 Synchronization of the 
event with the position of the mirror, 
although not essential in the Bowen 
RC-3 camera, was accomplished as 
shown in Fig. 3 and as previously de- 
scribed. 1 '* This synchronization was 
found to be extremely effective in 
improving the quality of the records of 
detonating charges. 

The rotating-mirror streak-camera is 
a powerful tool for the determination of 
velocity in explosive research, but 
caution must be taken in extrapolating 
information from the slit to other por- 
tions of the charge to which the camera 
is blind. For most explosive studies 
the low effective aperture of this type 
of camera is no handicap. High writing 
speeds are obtained with simple me- 
chanical systems by employing a long 
optical arm. 

In the Bowen camera the image speed 
along the film will resolve time to the 
accuracy required in the study of explo- 
sive velocities. However, the actual 
instantaneous shape of the detonation 
front, the air shock associated with it, 
and the change in shape of the detona- 
tion and shock fronts as the explosion 
progresses are not recorded by the 
streak-camera. 

Single-Exposure High-Speed Shutter (Rapa- 
tronic). A 4-/xsec Rapatronic Shutter,* 
which embodies the Faraday magneto- 
optic effect, was successfully perfected 
by the firm of Edgerton, Germeshausen 
and Grier of Boston. Dr. H. E. Edger- 
ton, in his many visits to our laboratory, 
expressed interest in the problems 
associated with the study of explosive 



Morton Sultanoff: Study of Explosives 



151 



Camera 



Fblarizing 
analyzer 




Po 



Source First 
Polarizer 



Glass slug 
Pulsing coil 



L_LJ 

E(impressed voltage) 
Fig. 4. Diagrammatic presentation of the principle of the Faraday shutter. 



reactions, and the experimental 1-jusec 
Faraday shutter was developed and 
loaned to our laboratory to be used in 
the study of explosives. An analysis 
of the applicability of that shutter to 
explosive studies has been made. 56 

In 1845 Faraday noted that a block 
of glass in a strong magnetic field dis- 
played optical activity. Further, he 
observed that when a beam of plane- 
polarized light passed through the glass 
parallel to the lines of force in the 
magnetic field the plane of polarization 
was rotated in accordance with: 



VHl 



where : 



4 is the angle of rotation in minutes of 

arc; 
V is the Verdet constant in minutes of 

arc/Gauss/cm (0.064 for the sodium 

5893 A line); 

H is the field strength in Gauss; and 
/ is the length of the path through the 

field in cm. 

The application of this effect to the 
magnetooptic shutter is shown in Fig. 4. 
The rays from the light source, L, are 
propagating in the direction D with 



random vibration. The polarizing 
screen P passes only the components of 
vibration shown. After the beam of 
plane-polarized light passes through the 
glass slug, which is normally optically 
isotropic, it is stopped by the polarizing 
analyzer P A whose axis is perpendicular 
to the original polarizer PQ. However, 
when a high-voltage pulse E is sent 
through the coil of wire which is wrapped 
around the glass slug, a rotation of the 
plane of polarization of the beam from 
PO results (Faraday effect) so that the 
analyzer PA is no longer perpendicular 
to the plane of vibration incident upon 
it. The components of the rotated 
beam P < in P A pass through to the 
camera. The duration of the pulse E 
therefore controls the duration of the 
passage of light through the analyzer 
PA to the camera. 

The main units of the Faraday shutter 
can be seen in Fig. 5. It is essential that 
the opening of the shutter be synchro- 
nized with the event so that a study may 
be made of an explosive reaction at any 
desired instant. To accomplish this 
synchronization to within 1 /xsec the 
Rapatronic shutter is supplied with a 



152 



February 1953 Journal of the SMPTE Vol. 60 




Discharge condenser 
and trigger gap 
unit 



Fig. 5. Faraday shutter: top, lens and slug; bottom, assembled 
shutter in 4 X 5 in. camera. 



Morton Sultanoff: Study of Explosives 



153 




Fig. 6. "Static" single exposure of spherical charge 
in firing position to show line structure. 



photoelectric pickup which receives a 
signal from the first light radiated from 
the event. This signal is then fed 
through a variable time delay which 
pulses the shutter at any selected time 
after the first burst of light. 

A large amount of light is lost in this 
shutter through polarization and ab- 
sorption in the glass slug. The physical 
size of the slug further reduces the 
aperture of the system. This light loss is 
tolerable since the luminosity of most 
explosive phenomena is very high. 

The Faraday unit is highly portable, 
simple to operate, and has given con- 
tinuous trouble-free service. In these 



respects it is superior to similar 1-jusec 
Kerr cell shutters which operate on the 
principle of birefringence displayed by 
nitrobenzene under an impressed electro- 
static field. 

High-Speed Motion Pictures (Grid Fram- 
ing Camera). The ultra-high-speed 
camera, which was developed specifi- 
cally for use in the study of explosives, 
has been described in considerable detail 
in an earlier Journal.* However, as a 
matter of review, the principle of 
multislit focal-plane scanning 7 on which 
this camera is based is described below. 

An image of the event to be studied 



154 



February 1953 Journal of the SMPTE Vol. 60 



Image on film plane 



Rotating 
mirror 




:rir_"iiiirr."_T: 

-Object 



First 
lens 




^Second 
lens 



Fig. 7. Schematic diagram of the optical system of the 
100,000,000 frame/sec camera. 



is formed through the multislit focal- 
plane shutter which is in intimate con- 
tact (optically) with the photographic 
emulsion. The parallel slits in this 
shutter are 0.0002 in. wide and are 
separated by opaque spaces of 0.020 in. 
A single exposure of a spherical explosive 
charge taken through this shutter is 
shown in Fig. 6. It should be noted 
that this picture resembles the image on 
a television screen in that it covers the 
full picture size, but only a fraction of 
the total area is involved in the image 
formation (1% for the shutter described 
above). If the shutter is now moved 
one slit width the first exposed image is 
completely covered and unexposed film, 
on which a second picture can be 
formed, is uncovered by the narrow slits. 
One hundred independent exposures can 
be formed with only 0.020 in. of motion 
of the shutter described. 

In order to photograph an event at 
the rate of 100,000,000 frames/sec the 
shutter is required to move 20,000 
in./sec. Since it is not feasible to move 
the shutter mechanically at this speed 
and still hold it in intimate contact with 
the film, as required to overcome the 
deleterious effects of optical diffraction, 
a rotating-mirror system as shown in 
Fig. 7 was employed. With this arrange- 
ment a 20-in. optical arm and a mirror 
velocity of approximately 100 rps pro- 
duces 100,000,000 frames/sec. An ac- 
curate 1 : 1 image of the grid is formed 
at the film plane so that when the corn- 





Fig. 8. Exposed plate of shock from 
pentolite sphere taken with the ultra- 
high-speed camera showing complete 
exposure and selected single frame. 

plete record of superposed images is 
obtained with this camera a single frame 
can be selected by the proper placement 
of the multislit shutter back on the 
exposed plate as shown in Fig. 8. 



Morton Sultanoff: Study of Explosives 



155 






S 
Q. 

x 



0> 



> 








February 1953 Journal of the SMPTE VoL 60 




" -s 

3 * 
& J 




O 

u_ 



O) 

C 

O 



a 

V 

a. 

s 

1 




- 

: 



Morton SultanofF: Study of Explosives 



157 





158 



Fig. 10. Devices for analyzing ultra-high-speed camera records: 
top, single-frame selector; bottom, motorized motion picture viewer. 

February 1953 Journal of the SMPTE Vol. 60 



This camera exhibits a very low light- 
gathering power and has a total run of 
only 100 independent frames. At the 
speed of 100,000,000 frames/sec a 
1-jusec run will expose the entire film 
area. It should be noted, however, 
that the events remain luminous for a 
considerably longer time than 1 /zsec, 
and repeated exposures are formed on 
each frame. The usefulness of this 
multiple exposure can be seen in Fig. 8, 
where about six images of the shock, 
each at 100-frame intervals, are ap- 
parent. A direct measure of velocity 
can be made from each independent 
frame, thus overcoming the errors which 
result from measurements that depend 
on data obtained from successive frames. 

Results of Photographic Studies 
of Explosive Mechanisms 

Detonation of Square Stick of Pentolite. 
The advantage of using all three types 
of basic cameras already described is 
displayed in Fig. 9. The photographs 
presented are of the reactions from a 
square stick of the standard explosive, 
pentolite. From the rotating-mirror 
streak-camera record the investigator 
can only obtain values for the velocities 
of the detonation front and the shock 
moving away from the end of the stick. 
Geometrical shapes of the explosive 
mechanisms are lost, and nothing can 
be accurately inferred about the action 
taking place in any part of the charge 
other than along the line which is in 
view of the camera slit. 

The Faraday shutter exposures (Fig. 
9c) show both side and end views of 
the detonating charge and reveal the 
shape of the luminous fronts and other 
luminous effects. However, only those 
events occurring at the instant of 
exposure are recorded. These pictures 
add a qualitative understanding to the 
quantitative data obtained with the 
streak-camera . 

The 100,000,000-frames/sec camera 
record shown presents a continuous 
study of the shape of the shock from the 



charge. Here continuous and corre- 
lated qualitative and quantitative data 
are obtained by studying the exposed 
plate played back as a motion picture, 
or by single frame, and frame-to-frame 
measurements in the viewers shown in 
Fig. 10. 

The accuracy of the velocities obtained 
are not as high with the ultra-high-speed 
camera as with the streak-camera, but 
knowledge is extended to include the 
whole charge. The quality of the 
exposure and therefore the discernible 
details in the ultra-high-speed camera 
picture are not as good as that of the 
Faraday shutter, but continuous studies 
must be obtained if information concern- 
ing the shapes of the explosive processes 
which change with time is to be studied. 

Shock From Pentolite Hemispheres and 
Spheres. Figures 11 A and 11B compare 
typical records from the three types of 
basic cameras. As with the square 
stick, each camera resolves those features 
for which it was intended, and a study by 
all three of the cameras employed at the 
Terminal Ballistic Laboratory is required 
to make an adequate analysis of the 
mechanisms associated with the detona- 
tion of a hemispherically shaped charge 
of pentolite. Each picture in the series 
of Faraday exposures was obtained from 
a separate charge with the shutter 
openings timed at the intervals shown. 

An extension of the streak-camera 
technique to include backlighted spheri- 
cal charges is shown in Fig. 12. 

Shaped (Munroe Effect) Charge. A com- 
plete understanding of the collapse of 
the cone, the formation of the jet, and 
the penetration of the target materials 
by the shaped charge has eluded the 
investigator in this field. A great deal 
has been learned about these mechanisms 
from photographic studies. The appli- 
cation of the three basic types of cameras 
to the study of shaped charges has resulted 
in the photographs shown in Fig. 13. 



Morton Sultanoff: Study of Explosives 



159 



Time 



Hemispherical 
charge 




a. "Rate" streak record 




(Multiple exposure) 
b. 



c. 



Fig. 11 A. Comparison of photographs of an exploding hemispherical charge 
taken with three "basic" types of cameras: a. rota ting-mirror streak-camera; b. 
ultra-high-speed motion picture camera; and c. single-exposure Faraday camera. 



160 



February 1953 Journal of the SMPTE Vol. 60 




Morton Sultanoff: Study of Explosives 



161 




be 





bo 

U 
- 

J5 
u 



aOUDjSIQ 




162 



February 1953 Journal of the SMPTE Vol.60 






Time 




Detonotion front 



Shock from 
collapsing cone 



Shock from 
end of charge 

First appearance 
of jet 





Direct photograph 



Backlighted 



b. 



Multiple exposure 
C. 



Fig. 13. Shaped charge photographs obtained with the three "basic" types of 
cameras: a. streak-camera; b. single Faraday shutter exposure; and c. ultra-high- 
speed camera record. 



In these studies it was noted that 
many of the events are not self-luminous. 
A source of backlighting was introduced, 
and by employing a semishadow tech- 
nique the photographs shown in Figs. 
13b and 14 were produced. 

Summary 

The material presented in this paper 
has been kept brief intentionally. How- 
ever, it is felt that sufficient reference 
is given to other publications to assist 



in a further study of the subject of photo- 
graphic instrumentation in the field of 
explosives. 

Synchronization of the various 
cameras is of considerable importance 
in many of these studies. This subject 
has been dealt with more completely 
in Ref. 8, which also contains a thorough 
study of the backlighting techniques 
which were developed at this Laboratory 
and which are mentioned briefly in 
this report. 



Morton Sultanoff : Study of Explosives 



163 




Fig. 14. Backlighted l-/usec Faraday shutter exposure of target 
penetration by shaped charge jet. 



In carrying out photographic studies 
of explosive reactions extreme caution 
is essential in the interpretation of the 
results obtained with any one type of 
camera. Employing the three types of 
cameras described in this report, the 
certainty as to the identity and nature 
of the explosive events is increased, but 
real scientific rigor cannot yet be as- 
signed to theories which have been 
postulated on photographic studies alone. 

Past British work in this field is out- 



standing, and in many cases is ahead 
of our work. The image converter 
(Table I, d. and m.) which is probably 
the most versatile and powerful single 
tool for the studies of detonation and 
shock is a good example of recent British 
development. 

With each phase in the study of the 
physics of explosions the researcher is 
faced with new problems and a require- 
ment for almost "fantastic" recording 
instruments. It appears that the only 



164 



February 1953 Journal of the SMPTE Vol.60 



limit imposed on speed, exposure time 
and light-gathering power of very- 
high-speed cameras is the imagination 
of the investigator. A brief survey of 
the references listed below the chart in 
Table I will make it obvious that this 
limit has proved to be only a minor 
handicap in pushing toward better 
cameras for the study of explosives. 

References 

1. (a) H. Muraour and A. Michel-Levy, 
Compt. rend., 198: 825, 1499, 1760, 2091, 
Jan.- June 1934. 

(b) H. Muraour and A. Michel-Levy, 
J. phys. radium, Ser. 7, Vol. 6: 496-498, 
1935. 

(c) S. Patterson, "Sources of light re- 
corded in photographs of detonating 
explosives," Nature, 167, No. 4247: 
479-481, Mar. 1951. 

(d) R. A. Bailey, D. R. Born and M. 
Sultanoff, "An optical study of shaped 
charge jets," Ballistic Research Labora- 
tories Report No. 788, Dec. 1951. Ob- 
tainable by authorized persons from 
Ballistic Research Lab., Aberdeen Prov- 
ing Ground, Aberdeen, Md. 

2. (a) Staff Writers, Life Magazine: 67-70, 
Oct. 23, 1950. 

(b) E. M. Pugh, R. V. Heine-Geldern, 
S. Forner and E. C. Mutschler, "Kerr 
cell photography of high speed phe- 
nomena," J. Appl. Phys., 22: 487-493, 
Apr. 1951. 

(c) E. M. Pugh, R. J. Eichelberger and 
N. Rostoker, "Theory of jet formation 
by charges with lined conical cavities," 
J. Appl. Phys., 23: 532-536, May 1952. 

(d) R. J. Eichelberger and E. M. 



Pugh, "Experimental verification of the 
theory of jet formation by charges with 
lined conical cavities," J. Appl. Phys., 
23: 537-542, May 1952. 

3. (a) Mallard and LeChatelier, Ann. 
mines (French), 4: 313, 1883. 

(b) H. B. Dixon, "On the movement of 
the flame in the explosion of gases," 
Trans. Roy. Soc. (London) A, 200: 
315-352, 1903. 

4. I. S. Bowen, "The C.I.T. rotating mirror 
camera, Model 2," OSRD Contract 
OE Msr-478, Apr. 27, 1945. Available 
from the California Institute of Tech- 
nology, Pasadena, Calif. 

5. (a) H. E. Edgerton and C. W. Wyckoff, 
"A rapid-action shutter with no moving 
parts," Jour. SMPTE, 56: 398-406 
Apr. 1951. (b) M. Sultanoff and R. A. 
Bailey, "The application of the Faraday 
magneto-optic effect to the optical 
study of explosive and shaped charge 
mechanisms," Ballistic Research Labora- 
tories Report No. 791, Nov. 1951. Ob- 
tainable by authorized persons from 
Ballistic Research Lab., Aberdeen Prov- 
ing Ground, Aberdeen, Md. 

6. M. Sultanoff, "A 100,000,000 frame per 
second camera," Jour. SMPTE, 55: 
158-166, Aug. 1950. 

7. F. E. Tuttle, "Improvements in high- 
speed motion pictures by multiple- 
aperture focal-plane scanners," Jour. 
SMPE, 53: 462-468, Nov. 1949. 

8. R. Holtzworth and D. Hinz, "Exploding 
wire backlighting for the study of 
detonation, shock and shaped charges," 
Ballistic Research Laboratories Report No. 
818, June 1952. Obtainable by au- 
thorized persons from Ballistic Re- 
search Lab., Aberdeen Proving Ground, 
Aberdeen, Md. 



Morton Sultanoff: Study of Explosives 



165 



Television Camera Equipment 
of Advanced Design 



By L. L. POURCIAU 



The servo circuits and mechanical design features of a television camera chain 
are described. The chain, in order to achieve flexibility and ease of opera- 
tion, features remote control of lens selection and focus. Centralized func- 
tional controls are grouped together for operating convenience. The equip- 
ment is compact, light in weight and sturdy for field use, yet meets the high 
performance standards required in the studio. 



I 



N THE COURSE of design of the camera 
chain and associated equipment to be 
described, major emphasis was placed on 
producing a set of equipment which 
would allow the television broadcaster 
to present to the viewer a picture of the 
highest quality at the lowest possible 
operating cost and would require a 
minimum of operator skill. 

Toward this end considerable atten- 
tion was given to functional design con- 
siderations. Controls were divided into 
functional groupings and placed accord- 
ingly. Control functions were stabilized 
to minimize the need for operator atten- 
tion during programming, and probably 
most important, control over lens aper- 
ture was brought to the camera control 
unit. In addition, provision has been 

Presented on October 6, 1952, at the 
Society's Convention at Washington, D.C., 
by L. L. Pourciau, General Precision 
Laboratory Inc., Pleasantville, N.Y. The 
substance of this paper was also presented 
orally at the 6th Annual NARTB Confer- 
ence, Chicago, 1 952. 



made for remote control of lens turret 
and optical focus, allowing cameras to be 
employed in positions where operation by 
a camerman is impossible or impractical. 

The equipment to be described is 
illustrated in Fig. 1. At the upper left 
is the camera; below, from left to right, 
are the camera chain power supply, 
master monitor, camera control unit, 
video switcher, and synchronizing signal 
generator. 

Some of the most interesting and novel 
features are to be found in the camera. 
Physically, it consists of a main frame 
plus a number of subassemblies, as 
shown in Fig. 2. The subassemblies 
may be removed for replacement or re- 
pair in a matter of minutes. 

Figure 3 is a block diagram of the 
camera. Neglecting the viewfinder for 
the moment, it may be seen that the 
camera contains the image-orthicon 
horizontal-sweep circuit, a sweep-loss 
protection circuit, a regulator for the 
focus-coil current, a pulse-type supply 
for the multiplier and image sections, 



166 



February 1953 Journal of the SMPTE Vol. 60 



Fig. 1. Camera chain (camera, power supply, master monitor, control unit, video 
switcher and synchronizing pulse generator). 




Fig. 2. Camera main frame and subassemblies. 
L. L. Pourciau: Television Camera 



167 




Fig. 3. Camera block diagram. 



and a servomechanism for optical focus 
control. In addition, there is a drive 
motor for the lens turret with the associ- 
ated actuating pushbuttons and control 
microswitches and the iris drive mecha- 
nism. 

The image-orthicon horizontal-sweep 
circuit is a conventional circuit with re- 
finements allowing linearity of better 
than 1% to be obtained. The sweep 
feed to the deflection yoke is balanced to 
reduce radiation of the flyback pulse. 
Pulses taken directly from the horizontal 
and vertical deflection coils serve to 
actuate a protection circuit which biases 
off the image orthicon in the event of 
failure of either or both sweeps. These 
same pulses are amplified and mixed to 
form the target-blanking waveform. 

The focus-coil current regulator acts 
to keep the current in the image-orthicon 
focus coil constant within 0.1% for a 10% 
change in coil resistance. This serves to 
prevent drifts in focus, sweep size, etc., 



which might otherwise occur. A circuit 
diagram of this current regulator is 
shown in Fig. 4. 

The current in the focus coil is 
measured by means of the 100-ohm re- 
sistor in series with the coil. A change 
in current through the coil causes a 
corresponding change in voltage drop 
across this resistor. This change in 
voltage is amplified and applied to the 
grid of the control tube in such a manner 
as to counteract the original change. 

The voltages required by the image 
and multiplier sections of the camera 
tube are supplied by rectifiers driven 
from a pulse-type supply which is syn- 
chronized at horizontal rate. 

The lens turret drive is quite novel in 
that it is electrically rather than manually 
controlled. Lens selection is achieved 
by means of pushbuttons on the rear of 
the camera. A set of eight micro- 
switches, actuated by cams on the rear 
of the turret plate, serve to stop the turret 



168 



February 1953 Journal of the SMPTE Vol. 60 




380 



100-"- 



Fig. 4. Image-orthicon focus coil current regulating circuit. 






RESISTORS IN 
H-ENS MOUNT 






TRAVERSE 
RATIO ^ 



e> 



!> 


if r 


~I 
1 

1 
1 


! FEEDBACK 
St / RATIO 


I / 1 I 


MOTOR 


4/^ ]<3w 0i | 6 ~ 


__| ERROR AMPLIFIE 


R GENERATOR 


CONTROl 
POT 

Fig. 5. 


POT. | 


r 

Focus servomechanism sche 


1 

1 
I 
1 


IMAGE ORTHICON 

CARRIAGE 
matic. 



when the selected lens comes into position 
and to cause the turret to take the short- 
est path to the selected lens. A spring- 
loaded magnetic brake is used to ensure 
positional accuracy and stability. The 
motor-driven turret provides positive 
and accurate lens indexing and offers the 
possibility of remoting the turret control 
if desired. 

The turret is supported at its pe- 
riphery in a ball race, which provides an 
extremely stable suspension and thus 
enables a large turret diameter to be 
employed. The turret lens circle is 
approximately 8 in. in diameter and will 



accommodate a very wide range of lens 
focal lengths. A hole extends the 
length of the camera from the center of 
the turret to allow special optical controls 
to be brought to the rear of the camera. 

Behind the turret is located a filter disk 
with accommodations for four filters. 
The camera is normally equipped with 
three : a minus blue, and two of neutral 
density, one with 10%, the other with 1 % 
transmission. 

The minus blue filter improves the 
color response of the 5820 or 5826 image 
orthicons, while the neutral density 
filters are needed to enable proper opera- 



L. L. Pourciau: Television Camera 



169 







CW 

ccw 



RE Jr* YS CONTROL SWITCH 

AT CAMERA 

a ecu- 



Fig. 6. Iris drive and indicator schematic. 



tion of the image orthicon where light 
levels are very high and also to allow 
lenses to be operated at more nearly 
optimum apertures. In addition, they 
enable reduction of depth of field where 
the director desires to concentrate audi- 
ence attention on a particular plane of 
interest. 

Control of optical focus is by means of 
a servomechanism. A simplified sche- 
matic of the servomechanism is shown in 
Fig. 5. The position of the image - 
orthicon carriage is controlled by a 
bridge arrangement in which an unbal- 
ance causes an error voltage to appear at 
the amplifier input, in turn causing the 
motor to rotate the error potentiometer 
in the direction which causes a reduction 
of the unbalance or error voltage. A 
generator on the motor shaft provides a 
feedback signal proportional to velocity 
to stabilize the system. The sensitivity 
of the servomechanism is such that the 
camera tube position can be controlled 
to within 0.001 in. 

A resistor which determines the ratio 
of carriage motion to focus control-knob 
motion is contained in the lens mount. 
This allows the ratio to be made propor- 
tional to the focal length of the lens in 
use. The ratio chosen allows focusing 
from infinity to close-up with one turn 
of the focus control knob for all lenses 
within the limits imposed by the 2\ in. 
of available carriage motion. Close-up 
is defined in this case as a diagonal object 
size of 9 in. This provides the camera 



operator with a focus control which 
"feels" the same regardless of the lens in 
use. Lacking this provision, the camera 
operator would have to compensate for the 
fact that the motion of the focus control 
would vary over a range of perhaps 8 or 10 
to 1 with the range of lenses commonly 
used. For especially close shots, the 
compensating arrangement may be dis- 
abled by a switch allowing full travel for 
any lens. 

In addition to the advantages offered 
by the ratio variation, servo control of 
focus allows for remoting of the optical 
focus control if desired. This feature, 
combined with the remote turret-posi- 
tioning control, allows placement of 
cameras in locations which, because of 
inaccessibility or other reasons, cannot be 
occupied by the cameraman. Control 
of optical focus and lens selection may 
then be brought to the camera control or 
director's position. 

Among the various image-orthicon 
controls which determine the quality of 
picture transmitted, it has been found 
that the lens aperture is one of the most 
important. In fact, other controls be- 
come more or less pre-set if convenient 
control over lens aperture is available. 
For this reason, the camera control 
operator has been provided with control 
over the camera lens aperture. This is 
accomplished by means of a motor- 
driven gear which meshes with a sector 
gear on the lens mount attached to the 
lens iris control ring. Motor rotation is 



170 



February 1953 Journal of the SMPTE Vol. 60 




Fig. 7. Camera control block diagram. 



VIDEO FROM 




MIXER OR TRANS. 



Fig. 8. Camera control unit video chain block diagram. 



controlled by means of a single lead in 
the camera cable. Control is, of course, 
available at the camera but is not 
ordinarily used. 

The position of the lens iris is indicated 
on meters calibrated in // numbers at 
both the camera and camera control 



unit. This is accomplished by means 
of a cam on the lens mount which is 
followed by a cam follower at the center 
of the turret. The cam follower in turn 
drives a potentiometer which provides 
the meter indication. 

Figure 6 is a schematic of the iris con- 



L. L. Pourciau: Television Camera 



171 



trol and indication means. The control 
relays are both normally closed and, con- 
sequently, no power is applied to the 
motor. If either relay is shorted by 
means of the switch, the relay opens and 
causes motor power to be applied. The 
motor rotates in a clockwise direction if 
the upper relay is shorted, or in a coun- 
terclockwise direction if the lower relay 
is shorted. Since both B-f and ground 
must be brought to the camera for other 
purposes, the only additional lead re- 
quired to bring control to the camera 
control unit is the one common to both 
relays. The iris-indicating meters read 
the voltage from the arm of the indicator 
potentiometer to ground. 

The camera viewfinder is essentially a 
high-quality picture monitor. The 
sweep circuits provide excellent linearity, 
and a pulse-type high-voltage supply 
synchronized at horizontal rate provides 
a source of accelerating potential which 
is independent of the horizontal-sweep 
circuit. The grid of the picture tube is 
clamped, providing excellent black-level 
control. A type 5FP4A picture tube is 
used in conjunction with a magnifying 
lens which serves not only to provide a 
larger picture, but to allow more com- 
fortable viewing on the part of the oper- 
ator. 

A block diagram of the camera control 
unit is shown in Fig. 7. Besides the 
main video amplifier chain, it contains 
picture and waveform monitors with 
associated circuitry, the image-orthicon 
vertical-sweep circuit, and the intercom- 
munication amplifier. 

A block diagram of the video amplifier 
chain is shown in Fig. 8. The first 
stage is a straightforward video amplifier. 
The second stage provides compensation 
for high-frequency losses in the camera 
cable. Variation in compensation is 
accomplished with a six-position switch, 
each position corresponding to 200 ft of 
cable, enabling compensation for up to 
1000 ft of camera cable. A second sec- 
tion of this switch causes a 10% increase 
in the a-c voltage supplied to the camera 



cable when cable lengths of 600 ft or 
longer are used. 

The third stage provides control of 
video gain by means of bias control of a 
remote cutoff tube. The output of this 
stage is clamped to remove hum which 
may have been picked up on the signal 
and also to provide a black-level refer- 
ence for the insertion of blanking and for 
white-peak clipper action. The clamp 
pulse is flattopped and of about 2 /xsec 
duration. It is derived from the com- 
plete synchronizing signal by means of a 
shorted delay line differentiating circuit. 
This allows the clamping signal to be set 
in the proper phase relationship with the 
incoming camera signal which may be 
delayed up to 3.3 /xsec, depending upon 
camera cable length. Control over the 
delay of this pulse is effected by means of 
the second section on the cable com- 
pensation switch. The clamp pulse has 
been made as wide as possible to elimi- 
nate the random streaking effects caused 
by clamping on noise peaks. 

Blanking and vertical shading are 
added to the signal at the plate of the 
clamped stage. Insertion of blanking in 
the plate of the clamped tube ensures 
constancy of "setup" with picture-con- 
tent variations. Because of this, there is 
usually no need to touch the "setup" 
control during a program. Following 
blanking insertion, the signal is clipped 
and fed to the peak-white limiter. The 
peak-white limiter helps to prevent ex- 
cessive modulation of the transmitter or 
overloading of external video line am- 
plifiers. The control is usually set to 
clip peaks which exceed 1.1 v for a nor- 
mal white level of 1.0 v. After passing 
through the peak-white limiter, syn- 
chronizing signal is added to the video 
signal if desired, and the complete signal 
is fed to the video output stage which 
feeds the 75-ohm output line. 

Both picture and waveform monitors 
are normally fed from the output line. 
However, a "transmit" relay is provided 
which allows the output video to be cut 
off, in which case the pick-off point for 



172 



February 1953 Journal of the SMPTE Vol. 60 



Fig. 9. 

Camera control unit. 




the monitors is switched to a point 
ahead of the output stage. This switch 
can be used in case of trouble at the cam- 
era to prevent the switcher operator 
from putting the camera on the air. It 
may also be used with a ready-light sys- 
tem to provide the director with an 
indication of camera availability. 

The camera vertical sweep circuit is 
contained in the camera control unit, 
reducing the circuitry required in the 
camera and at the same time making the 
camera vertical-sweep control available 
to the camera control operator. The 
sweep circuit employs sufficient feedback 
to eliminate the need for a vertical 
linearity control. The feedback main- 
tains vertical linearity to well within 
1%. 

Control over camera horizontal-sweep 
amplitude is also available at the camera 
control unit as is an "overscan" switch 



which provides a 5% increase in camera 
sweep amplitudes. The overscan switch 
makes it quite simple to overscan the 
target during rehearsal, thus effectively 
increasing the useful life of the image 
orthicon. 

The picture monitor uses a type 
8AP4A cathode-ray tube. The type 
8AP4A was chosen because of its light 
weight and its high ratio of useful picture 
area to tube size. It presents a picture 
only 1 5% smaller than that presented by 
a 10-in. tube, yet is 2 in. smaller in diam- 
eter, thus providing an excellent com- 
promise between the needs of studio and 
field uses. 

Picture monitor vertical- and horizon- 
tal-sweep circuits provide linearity of 
better than 1%. The vertical circuit 
employs feedback to provide excellent 
long-term stability. A separate pulse type 
high-voltage supply, identical with that 



L. L. Pourciau: Television Camera 



173 




Fig. 10. Camera chain power supply 




174 



Fig. 11. Synchronizing pulse generator. 
February 1953 Journal of the SMPTE Vol.60 



I used in the viewfinder, provides 8 kv with 
[ good regulation for the 8AP4A anode. 

The waveform monitor employs a type 
j 3KP1 cathode-ray tube in a circuit 
; which provides excellent focus over the 
| tube face. Three choices of sweep speed 
are provided by means of a switch on the 
front panel. They are: half-line rate, 
half-field rate, and a sweep at field rate of 
approximately twelve lines' duration. 
This latter may be used to examine the 
vertical synchronizing signal as a check 
on synchronizing signal generator opera- 
tion. 

A calibration means is provided and 
the calibration may be conveniently 
checked by depressing the "calibrate" 
button on the front panel. The wave- 
form monitor may also be used as a test 
oscilloscope by means of a jack on the 
rear panel. An edge-lighted standard 
IRE scale is provided in conjunction 
with a green filter, matched to the spec- 
tral characteristic of the PI phosphor, 
which provides a presentation of excellent 
contrast. 

To allow single camera intercom- 
munication operation independent of 
external equipment, an intercommunica- 
tion amplifier is provided in conjunction 
with an intercommunication system, 
offering great flexibility. A three-posi- 
tion key switch on the front panel 
allows three modes of system operation. 
With the key in the upper position, all 
intercommunication positions have full 
privileges, that is, all may talk and may 
listen. In the center position, only the 
director has full privileges and all other 
stations may listen only. The lower 
position provides a private line between 
camera and camera control unit to en- 
able the operators to converse without 
disturbing the main intercommunication 
line. This is particularly useful if a 
camera is in trouble and cooperation be- 
tween video operator and cameraman is 
necessary for rectification. A call but- 
ton is provided at the camera which 
causes a light to flash at the camera con- 
trol unit enabling the cameraman to 



attract the video operator's attention 
when necessary. In operation the upper 
switch position would be used during re- 
hearsals, and the center position would 
be used while on the air to prevent 
interference with the director's instruc- 
tions. 

Figure 9 serves to illustrate the func- 
tional grouping of the controls on the 
camera control unit. The large knobs 
in the lower left- and righthand corners 
are the setup and video gain controls, 
respectively. The edge controls, be- 
tween the large knobs, control target 
voltage and beam current. The push- 
button at the lower left provides a nega- 
tive 2-v shift in target voltage to allow 
the target to be easily set to 2 v above 
cut-off. This provides a convenient 
"stepping-off" point for final target- 
voltage adjustment since optimum target 
potential will usually be found to lie be- 
tween 1.8 and 2.2 v above cutoff. 

The three small knobs just above the 
gain control, reading from left to right, 
are the beam-focus, image-focus, and 
multi-focus controls. The pushbutton 
to the left of these controls provides a 
check on waveform-monitor calibration. 
The key switch to the right of the wave- 
form monitor controls the waveform- 
monitor sweep rate. The "transmit" 
switch is to the far right, with the iris 
indicating meter in between these two 
switches. Just above the transmit switch 
is the iris control key, and in the corre- 
sponding position on the left is the inter- 
communication key. Knobs at the top 
left and right control the picture-monitor 
contrast and brightness. Other con- 
trols are found under the door on top of 
the unit and along a panel running down 
the left side. 

The camera chain power supply is 
shown in Fig. 10. It is only 12 in. 
high and weighs 87 Ib. Components are 
mounted on two standard rack panels, 
one of which swings out, as illustrated, to 
provide access to the undersides. A 
meter on the front panel allows measure- 
ment of all d-c voltages and currents in 



L. L. Pourciau: Television Camera 



175 



3I5KC 





rBLANKWGl 
COMPSYNC. 

' H HOR.DRIVE h 

[VERT.DRIVEj 



Fig. 12. Synchronizing pulse generator block diagram. 



MAIN DIVIDER 



3I.5KC 



60~ 




OUTPUT 


> T 22sfJ 


AF 


'C 
UIT 




315 K.C. 
MASTER OSC. 




I*** \ CHASS ' 5 


REF FREQ. 

J 

^ 


"* CIRC 




ENCY 


COUNTER ' 


j 

PHASING : 

r i 


f REACTANCE 
i TUBE 
BIAS 

r 


!MO. 
FREOU 


3I.5KC TO MAIN 


DIVIDER 



Fig. 13. Master oscillator block diagram. 



addition to the a-c line voltage. A line- 
tap switch allows operations over an in- 
put supply range of 95 to 130 v. The 
power consumption of a complete chain 
is approximately 1000 w. 

The synchronizing pulse generator is 
packaged as a completely self-contained 
unit. The circuits are built on standard 
rack panels and, when rack-mounted, 
the complete synchronizing pulse gener- 
ator with power supply requires only 31 
in. of rack space, yet operational sta- 
bility is such that only three controls 
need be provided. The synchronizing 
pulse generator, open for servicing, is 
shown in Fig. 1 1 . 

A simplified block diagram of the syn- 
chronizing pulse generator is shown in 
Fig. 12. The master oscillator operates 



at 31.5 kc. From this is derived a 
pulse at 60-cycle rate by means of the 
525 : 1 divider, and a pulse at 15.75 kc 
by means of the 2 : 1 divider. These two 
pulses are fed to the synchronizing and 
blanking pulse generator circuits and 
are used to generate the desired output 
waveforms. These are the complete 
synchronizing signal, mixed blanking 
signal, and horizontal and vertical drive 
pulses. All four waveforms are avail- 
able in either positive or negative polar- 
ity at a level of 4 v. 

A block diagram of the master oscil- 
lator chassis is shown in Fig. 13. A 
twin triode, 31.5-kc oscillator, which 
provides excellent stability, is used. An 
AFC circuit allows the oscillator to be 
locked to the 60-cycle power line or to an 



176 



February 1953 Journal of the SMPTE Vol.60 




Fig. 14. Bi-stable multivibrator. 

external 60-cycle reference. The chassis 
also contains a single binary counter 
stage which provides 15.75-kc pulses to 
the pulse former chassis. The master 
oscillator panel contains the only circuit 
controls in the synchronizing pulse gen- 
erator. They are: the frequency con- 
trol for the oscillator, a phasing control 
for adjusting the phase of the 60-cycle 
lock-in, and a bias control on the react- 
ance tube. The reactance-tube current 
may be read on the front panel meter. 
This enables the synchronizing pulse 
generator to be locked to the reference 
frequency at the center of the AFC con- 
trol range without the necessity of ex- 
ternal test equipment. 

The counter chassis consists of ten 
binary counter stages which are made to 
provide a count of 525 by means of pulse 
feedback. The chassis accepts 31.5-kc 
pulses from the master oscillator and 
furnishes a pulse at 60-cycle rate to the 
pulse-forming circuits. Other outputs 
at various points along the divider chain 
are also used in the pulse-former in the 
interest of overall simplification. The 
binary type counter is a very simple and 
very stable circuit requiring no adjust- 
ments whatsoever. 

The pulse-former chassis accepts out- 
puts from the master oscillator and 
divider chassis and generates the desired 
output waveforms. All timing of pulse 
widths, including the equalizing and 
vertical-synchronizing pulse-gating wave- 
forms, are determined by a delay line or 
by pulse counting by means of binary 



stages. Thus, once the desired pulse 
widths are chosen there is no need for 
further adjustment. 

A generalized form of the circuit used 
to generate the output pulses is shown in 
Fig. 14. It is seen to be a variation of 
the well-known Eccles- Jordan trigger 
circuit which is now known as a bi- 
stable multivibrator. This latter term is 
quite explicit since the circuit has two 
stable conditions: one in which tube A 
is fully conducting and tube B completely 
cut off; the other with tube B fully con- 
ducting and tube A cut off. It may be 
transferred from one condition to the 
other by a number of triggering means. 
For instance, if a trigger is ap- 
plied to a point which is common to 
both triodes, such as the cathode con- 
nection, a transition from one condition 
to the other occurs with each trigger 
pulse. If the circuit is to be used to 
generate a pulse of a particular duration, 
triggers corresponding in time to the 
leading and trailing edges of the required 
pulse may be applied respectively to the 
points "a" and "b". A transition from 
one stable state to the other occurs upon 
arrival of the leading-edge trigger and a 
return to the original stage occurs upon 
arrival of the trailing-edge trigger, thus 
causing the circuit to generate a pulse of a 
duration corresponding to the time inter- 
val between the leading- and trailing- 
edge triggers. 

The circuit is used in this manner to 
generate both horizontal and vertical 
rate pulses. In the case of the hori- 
zontal pulses, triggers of the correct 
timing are obtained from a delay line. 
Vertical-rate triggers are obtained from 
the main divider and from a secondary 
divider chain which operates in con- 
junction with the main divider. 

The synchronizing signal generator 
power supply furnishes unregulated 200 v 
to all circuits except the pulse clippers 
and the master oscillator. The voltages 
for these circuits are regulated by means 
of a pair of VR tubes. The a-c power 
consumption is 250 w. 



L. L. Pourciau: Television Camera 



177 




Fig. 15. Video switcher. 



The video mixing unit, or switcher, is 
illustrated in Fig. 15. It is designed to 
provide studio switching facilities in a 
field package. The switch panel is 
shown in the operating position. When 
not in use or when in transit, it may be 
swung back into the main unit. The 
box containing the switching facilities 
may be easily removed and can be 
operated at distances up to five feet 
from the main unit. The switcher will 
accept inputs from as many as five local 
and two remote signal sources. A pre- 
view buss, two effects busses and an in- 
stantaneous switching buss are provided. 
The instantaneous buss and the two 
effects busses with their associated faders 
allow cuts, superimpositions. fades, etc. 
The preview buss provides for the pre- 
viewing of any input signal or the output 
of the effects busses, thus allowing an effect 
to be properly set up before transmission. 
A "transmit" button on this buss provides 
means for monitoring the output line. 
The main output provides a 75-ohm 
source impedance for proper matching 
to telephone company lines. 

A block diagram of the video mixing 



unit is shown in Fig. 16. It may be 
seen that black level clamping is pro- 
vided to eliminate switching transients. 
The output line is monitored directly, 
but through a pad arrangement which 
isolates the monitor from the hum and 
tilt usually introduced by a telephone 
company line connection. Besides the 
line and master monitor outputs, an 
effects monitor output and a separate 
line monitor output are provided. The 
effects monitor output is used to monitor 
the output of the effects busses, and the 
separate line monitor output may be 
used to drive a studio or announce mon- 
itor. 

The switcher intercommunication cir- 
cuits provide intercommunication fa- 
cilities for the director and allow trans- 
mission of program audio to other inter- 
communications positions if desired. 

The power supply for the video switch- 
ing unit is self-contained. 

The master monitor (Fig. 17) is de- 
signed to work in conjunction with the 
video switcher or as a general-purpose 
monitor. It includes both an 8^-in. 
picture monitor and a 3-in. waveform 



178 



February 1953 Journal of the SMPTE Vol.60 




L. L. Pourciau: Television Camera 



179 




Fig. 17. Master monitor. 



monitor. Facilities are provided for 
switching to one of three input signals. 
A synchronizing signal interlock arrange- 
ment is provided to enable the unit to 
handle both composite and noncom- 
posite signals. 

Picture and waveform monitor circuits 
are very similar to those used in the 
camera control unit. Accurate mainte- 
nance of black level is assured by means 
of a clamp on the cathode-ray tube grid. 

Three sweep rates are provided for the 
waveform monitor, corresponding to 
those in the camera control unit. 

The design and development of the 
equipment which has been described are 
the result of the efforts of a considerable 
number of engineers, both at General 



Precision Laboratory Incorporated and 
at Pye, Ltd., in Cambridge, England. 
Much credit goes to Messrs. J. E. Cope, 
L. W. Germany, and D. Jackson who 
were responsible for the work at Pye, Ltd. 

Discussion 

Barton Kr enter (RCA Victor Div., Camden, 
N.J.): Did I understand that this camera 
had a remotely controlled lens-selecting 
mechanism for changing lenses? As I 
understand it, it selects the shortest path 
for the incoming lens. The lens never has 
to travel as great an arc as the circle. 
What is the maximum time it would take 
to change the lens? 

Mr. Pourciau: I think it runs about 1 \ 
to 1^ seconds per lens, so that the maxi- 
mum time would be about 2^ to 3 seconds. 



180 



February 1953 Journal of the SMPTE Vol. 60 



Splicing Motion Picture Safety Film 
Without Cements or Adhesives 



By LEONARD A. HERZIG 



A method and apparatus for butt-weld splicing motion picture film, not re- 
quiring the usual scraping or cements, are described. The principle is 
based on a combination of a controlled heat and cooling gradient applied 
under pressure within a given time cycle, and producing a homogeneous 
splice. Properties of the film are not affected, as the film is automatically 
preplasticized prior to splicing. The method may be used for all types of 
safety film. 



JT ILM SPLICING has been a serious 
problem since the advent of motion 
pictures. The standard method of splic- 
ing, in use for the past decade, has 
required scraping, cementing and over- 
lapping the film. In some cases it has 
also required the application of heat to 
speed up the drying time of the solvent. 
The problem has been increased with 
the advent of different types of safety 
bases such as standard acetate, tri- 
acetate, butyrate-acetate, etc. Addi- 
tional difficulties are now forthcoming 
with the use of magnetic-striped sound- 
tracks and magnetic film, where an 
overlap splice introduces distortion and 
loss of sound. 

A new method of splicing has been 
developed which incorporates the prin- 
ciple of butt-welding film end-to-end 



Presented on October 7, 1952, at the 
Society's Convention at Washington, B.C., 
by Leonard A. Herzig, Prestoseal Mfg. 
Corp., 37-27 33rd St., Long Island City 1, 

N.Y. 



and eliminates the need of scraping, 
cementing and overlap. 

In order to obtain a satisfactory butt- 
weld splice, many factors had to be 
considered. One of these was to obtain 
a precise controlled heat having a heat 
gradient confined in area to very narrow 
limits. Another was to find a material 
with a cooling gradient capable of 
reducing to the required temperature 
within a maximum period of three 
seconds. Of these two factors, it was 
found that the cooling gradient was of 
prime importance and took precedence 
in the designing of the proper heater 
block. Materials which have very poor 
heat-conducting properties, but which 
would have a proper heat gradient as 
well, were required. It was necessary 
that the parts be able to withstand ex- 
treme, as well as sudden, changes in 
temperature. Also, as mentioned pre- 
viously, the heating element itself had 
to be capable of cooling in a period of a 
few seconds. 



February 1953 Journal of the SMPTE Vol. 60 



181 




Fig. 1. Heater block assembly. 

To accomplish this, it became neces- 
sary to discard all types of coil heating 
elements and to introduce a ribbon type 
of nichrome wire (Fig. 1 (6)). This 
ribbon has a cross-section area no 
greater than 0.006 X 0.047 in. and 
is capable of dissipating 40.5 watts 
per linear inch, or an equivalent of 
0.1142 Btu. The heater ribbon is sunk 
into a piece of transite material (Fig. 
1 (2)), until it is absolutely flush with 
the surface of the transite. The ac- 
curacy of the positioning of this ribbon 
wire in the transite is important. The 
nichrome ribbon carries a current of 
approximately 16 amp which, in air, 
would normally cause the ribbon to 
burn out. This current-carrying ca- 
pacity of the ribbon changed con- 
siderably when the wire was placed 
under pressure due to the dissipation of 
heat from the ribbon to other materials 
in pressure contact with the ribbon. 
To prevent undue burning out of the 
heater wire, safety switches had to be 
provided which would prevent the 
operator from splicing without first 
applying pressure to the film and 
heater element. 

Mica (Fig. 1 (3)) is used as one of 
the heating and cooling gradients. 
The type of mica used, as well as the 
thickness of the mica, is extremely 
important. This mica also serves as 



an electrical insulator to prevent con- 
tact shorting of the heater ribbon to the 
stainless-steel platen. The thickness of 
the mica determines the heating and 
cooling gradient. The optimum thick- 
ness has been found to be between 0.002 
and 0.003 in. 

In the development stage of the 
Presto-Splicer, mica was used as a 
heater platen, but it tended to flake 
and adhere to the film being spliced, 
which, of course, was not desirable. To 
overcome this, a stainless-steel platen 
(Fig. 1 (7)), 0.005 in. in thickness, 
having a high electrical resistivity, was 
placed over the mica. It is interesting 
to note that resistivity played an im- 
portant part, and, inasmuch as re- 
sistivity is inversely proportional to heat 
conduction, the use of a specific type of 
stainless steel became necessary. This 
steel introduced problems of warping 
and elongating under heat, so that it 
therefore became necessary to have the 
grain of the material perpendicular to 
the line of heat. 

Pressure Requirements 

After elaborate testing, it was found 
that pressure under 200 psi applied to 
the line of splice allowed gas bubbles 
to appear in the splice, causing a poor 
bond and brittleness of the film. When 
pressure of 200 psi or more was applied 
to the splicing area, these gas bubbles 
disappeared and a satisfactory homo- 
geneous bond resulted. 

Heat Transfer 

With the advent of tri-acetate film, 
it became necessary to find the material 
having the poorest heat-conducting 
characteristics which would allow the 
concentration of heat to be applied to 
the film rather than* to be dissipated to 
the material used for applying the 
pressure to the film (Fig. 2). Other 
requirements of this pressure platen 
were that it should in no way affect 
or adhere to the emulsion of the film. 
At present, the material used for this 



182 



February 1953 Journal of the SMPTE Vol. 60 



TEFLON PRESSURE 
PLATEN 



TEFLON TAPE 




Fig. 2. View of splicer showing Teflon platen and tape. 



pressure platen is "Teflon" (tetrafluoro- 
ethylene). The type of orientation of 
this Teflon block is important as Teflon 
has a very serious contraction and 
expansion characteristic. It also became 
necessary to restrain its mechanical 
motion by encasing the block with a 
U-shaped sleeve. One of the disad- 
vantages of this material is that it is 
extremely soft and tends to mark very 
easily. To overcome this, a roll of 
Teflon tape was placed between the 
Teflon block and the heater block. 
This tape automatically advances with 
every splice. 

Plasticizer 

When heat is applied to acetate 
film, there is a tendency for this heat 
to drive out a small proportion of the 
plasticizer from the film. This tends 
to make the splice brittle. To over- 
come the loss of plasticizer, an automatic 
plasticizing unit (Fig. 3), which plas- 
ticizes the edge of the film before it is 
spliced, had to be added to the splicer. 



This plasticizer had to handle all of 
the various film bases which are in 
existence today. In addition to pre- 
venting brittleness of the splice, the 
plasticizer also broadens the flow point 
of high-acetate film, which previously 
required precise heat control to within 
1 F. With the introduction of the 
plasticizer, the permissible temperature 
range has been broadened by 10 F. 

The Presto-Splicer, shown in Fig. 4, 
consists of a foundation base and an 
interchangeable 16mm or 35mm head. 
The base contains all the electrical and 
timing components with the exception 
of the heater element, located in the 
head assembly. The interchangeable 
heads can either be removed from the 
base or assembled thereto in about 30 
sec. Two knurled screws are used for 
interlocking the head to the foundation 
base. A tongue-and-groove locking de- 
vice is located on the front of the head 
and the base assembly in order to add 
additional rigidity. 



Leonard A. Herzig: Film Splicing 



183 



The overall dimensions of this splicer 
are: 

Height 9 in. 

Width lUin. 

Length 17 in. 

Weight 17ilb. 

Maximum power consumption .100 vv 

The splicer operates on 110/120 v, 
50/60 cycles and can be converted for 
use with 220 v. The castings through- 
out are #108 aluminum alloy with a 
hard-baked black crinkle finish. 

Splicing of Dissimilar 
Safety Base Materials 

No trouble has been encountered 
when splicing present types of film, 
whether to the same or to dissimilar 
film. However, different time and 
heat settings are necessary when splicing 
dissimilar materials to each other. 
These settings can be classified as low, 
medium and high. The Du Pont safety- 
base film falls into the low category 
as it has a low heat requirement. Nega- 
tive Eastman Kodak stock falls within 
the medium heat requirements, and posi- 
tive Eastman Kodak stock falls into 
the high heat requirements. When 
splicing a low heat requirement ma- 
terial to a high heat requirement ma- 
terial, the higher of the two settings is 
used. 

Accuracy of Indexing 

In the current method of cement 
splicing, accuracy of the cut with rela- 
tion to the sprocket hole does not 
introduce any serious problem. How- 
ever, where butt-welding is to be 
achieved, this cut, with relation to the 
sprocket hole, has to be kept very 
accurate. To accomplish this, it was 
necessary to take into account the 
0.2% shrinkage tolerated in the motion 
picture field. Obviously, this elimi- 
nated the possibility of using two or more 
sprocket indexing pins and it became 
necessary to use only one indexing pin 
(Fig. 3) for both cuts. 



Clamping and Parallel Alignment 

The film is clamped in the conven- 
tional way except that a banking edge 
is required in order to obtain perfect 
alignment control throughout the splic- 
ing cycle. The clamps, therefore, in- 
corpoj-ate a pair of film followers (Fig. 
3) which push the film to the banking 
side of the clamp prior to the clamp's 
being closed. 

After the film is indexed and clamped, 
it is then cut, and the clamp still holding 
the film is swung through a 180 arc 
into the heat-sealing position. The 
same cutting blade is used to cut both 
pieces of film. As the second clamp is 
rotated to the heat-sealing position, the 
edge of the film to be heat-sealed is 
coated with the plasticizer (Fig. 3). 

At this point it should be noted that 
one of the major splicing technique 
changes is that the film is clamped with 
the emulsion side facing down and the 
cellulose side up. Of course, when the 
film is pivoted over a 180 arc to the 
heat-sealing position, the emulsion is 
facing up and the pressure is applied to 
the emulsion surface. The heat is then 
applied to the cellulose side (Fig. 2). 

Acetate Flow During the Heat Cycle 

16mm and 35mm film each present 
different problems during the heat 
cycle. The frameline of a 1 6mm picture 
is on line with the sprocket hole, which 
fills up completely during the heat 
cycle. With 35mm film, there is also 
the tendency of the acetate to flow 
into the four sprocket holes adjacent 
to the point of weld. To prevent the 
flow of acetate into these sprocket holes, 
ears were stamped into the heater 
platen (Fig. 5). These ears are the 
same size and shape as the sprocket 
hole and extend above the stainless- 
steel platen by 0.006 in. The acetate, 
besides flowing sideways, would also 
flow along the line of heat and out each 
end if it were not restrained. To prevent 
this, edge-flow plates (Fig. 5) made of 
stainless steel are placed against the 



184 



February 1953 Journal of the SMPTE Vol.60 






FILM 
FOLLOWER. 



INDEX 
PIN 



PLASTiCtZER BANKING 
WICK WALL 




Fig. 3. Automatic plasticizing unit. 



HEAD 




BASE 



Fig. 4. The Presto-Splicer. 
Leonard A. Herzig: Film Splicing 



185 



banking edges of the film. The edge- 
flow plates must be slightly thicker 
than the film to prevent the flow from 
going over the top of the plates. Locking 
clamps (Fig. 1 (9)), to lock the edge- 
flow plates to the heater block, are 
also used. 

Splicing of Magnetic Film 

When splicing magnetic film, it is 
necessary to replace the stationary and 
movable knife blades with nonmagnetic 
materials, Fig. 5. The same procedure 
is followed for splicing magnetic film 
as that used for 16mm or 35mm motion 
picture film. One of the advantages of 
magnetic film over motion picture film 
is that the oxide coating does not tend 
to roll back the few thousandths of an 
inch which ordinarily seems to be 
characteristic in splicing motion picture 
film. 

When a machine is to be used strictly 
for 16mm magnetic film, it is advisable 
to change the location of the index pin 
so as to splice between sprocket holes 
rather than through them, as this 
eliminates the reperforating process. 
Obviously, there are very few users 
who could afford to tie up a splicing 
machine for work on magnetic film 
only. Therefore, this change is never 
incorporated unless specifically re- 
quested. 

Frequency tests run on magnetic film 
disclose the following: 

A film with no modulation was 
spliced every 20 ft and recorded on 
without any wiping or de-magnetizing 
of the film. No audible noises were 
noted. An additional film, having a 
frequency range of from 30 to 15,000 
cycles recorded on it, was spliced in 
each of the various fixed frequencies. 
This was checked on RCA equipment 
with the following results: 

Frequencies from 1,000 to 15,000 
cycles splice was inaudible or at 
the same level as that of the normal noise 
of the material. 

Frequencies from 700 to 1 ,000 cycles 



slightly noticeable when the magnetic 
film was run backward or forward in 
this range. 

Frequencies below 700 cycles in- 
audible. 

Strength tests indicated that the 
splic,ed area was equal to 90% of the 
tensile strength of the film itself. 

When 16mm or 35mm standard print 
stock with a stripe of magnetic material 
on the side opposite to that of the 
emulsion is used, it is necessary to change 
the stainless-steel heater platen (Fig. 1 
(7)), to a special stainless-steel, Teflon- 
coated platen. This prevents the oxide 
coating from sticking to the heater 
platen. 

Splicing of Color Film 

A number of companies are using 
this splicer for splicing color raw stock 
and, again, this Teflon-coated platen 
had to be used to prevent the anti- 
halation coating present on coated 
raw stock from sticking. When splicing 
color prints, it was found that replacing 
the Teflon tape with a cellophane tape 
was advantageous. 

Splicing of Negative Film 

Splicing of negative 16mm or 35mm 
film gives exceptionally gratifying results. 
When prints are made from spliced 
areas, no indication as to where the 
splice was made can be noted. No 
out-of-focus frames are introduced and 
perfect registration of the picture is 
achieved without the side shift normally 
observed as a result of the cement 
method of splicing. 

Splicing of Raw Stock 

By removing the viewing lights or 
placing a Wratten filter over them, it 
is possible to splice raw stock in the 
dark. Previously, splicing of raw stock 
entailed using the "hit-or-miss" method 
as to whether the emulsion was com- 
pletely scraped off, and, therefore, the 
splicing of raw stock was rarely at- 
tempted. Because in the butt-weld 



186 



February 1953 Journal of the SMPTE VoL 60 



LOCKING CLAMPS 
EDGE FLOW PLATE 



STATIONARY 
KNIFE BLADE 




Fig. 5. Details of Presto-Splicer. 



method no scraping or cements are 
used, it is relatively easy to splice raw 
stock without light. It is a decided 
advantage when splicing 16mm raw 
stock in this way to move the indexing 
pin to the center of the frame in order 
to avoid having to reperforate in the 
dark. This method of splicing has been 
successfully used in darkrooms for proc- 
essing, eliminating the need for stapling 
leader stock to "takes." It has also 
been used by a few of the larger com- 
panies for telefax work, where tempera- 
tures in the developing tank go as high 
as 125 F, thus avoiding the trouble 
encountered when rivets are used. 

Effect of Ambient Temperature and 
Relative Humidity on Splicing 

Temperatures ranging from 70 to 
80 F seemingly have no effect on the 
splicing. When the ambient tempera- 
ture rises above 80 F, it is necessary to 
reduce the current by 5%. This 5% 
is correct for all temperatures up to 
1 1 5 F. When the ambient temperature 



of the room ranges between 55 and 
68 F, the current is increased 5% over 
the normal setting. 

Relative humidity ranging from 40 to 
100% is exceptionally good for the 
butt-weld method. When the relative 
humidity is between 10 and 20%, it is 
necessary to keep the plasticizer wick 
very moist. 

Tensile and Flexing Strength 

Tensile strength of the butt-weld 
splice ranges between 90 and 95% of 
the film itself. The flexing strength 
has not been tested on a flexometer but 
has been placed on endless loops in all 
types of projectors. Tests have been 
run in excess of 1,000 times before any 
edges showed signs of cracking. 

Conclusion 

Butt-weld splicing has many ad- 
vantages over conventional cement splic- 
ing. Splices can be made every 10 sec 
and used immediately in the darkroom 
either for raw stock, negative, positive, 



Leonard A. Herzig: Film Splicing 



187 



color, dissimilar materials, magnetic in the microfilm field, where the loss of 

materials, striped stock or for the new document area is not tolerated, 

film, "Milar," recently announced by We believe that, now that the motion 

Du Pont. Prior to the introduction of picture industry has changed over to 

this equipment to the motion picture safety film, the Prestoseal butt-weld 

field, the Presto-Splicer had been time- method of splicing can be of invaluable 

tested by many Government agencies service to this industry as well. 



ASA Photographic Standards Board (PSB) 

IN ACCORDANCE with the recommendation of the Committee on Procedure of the 
American Standards Association, approved by the Standards Council on January 
14, 1953, the name of the Photographic Standards (Correlating) Committee (PS(C)C) 
has been officially changed to the Photographic Standards Board (PSB). 

The complete procedures in the development of American Standards were pub- 
lished in the August 1952 Journal, p. 155. 



Proposed American Standard PH22.75 

A and B Windings of 1 6mm Single-Perforated Film 

THE PROPOSED American Standard on A and B Windings of 1 6mm Single-Perforated 
Film was published previously in the September 1949 and January 1951 Journal 
and is again published on the following page for three-month trial and comment. 

A major manufacturer of equipment using a nonstandard winding took exception 
to the A winding on spools as specified in the January 1951 draft. Apparently the 
SMPE 1941 Recommendation upon which the proposal was based was misinterpreted 
and consequently this winding was reversed. Inasmuch as the demand for A-wound 
film on spools is small and supplied in the main by one firm, the 16mm and 8mm 
Motion Pictures Committee agreed to delete this specification entirely. 

If no adverse comments are received this proposal will be submitted to the ASA 
Sectional Committee PH22 without further balloting. H.K. 



188 February 1953 Journal of the SMPTE Vol.60 



Proposed American Standard 

A and B Windings of 16mm 
Single- Perforated Film 



(Third Draft) 



PH22.75 



The purpose of this standard is to insure a 
uniform method of designating the type of 
winding (the location of the perforated edge) 
when ordering or describing 16mm raw-stock 



Winding A 
Emulsion side in 



Film on Cores for Darkroom Loading 

When a roll of 16mm raw stock perforated 
albng one edge is held so that the outside end 
of the film leaves the roll at the top and 
toward the right, winding A shall have the 
perforations along the edge of the film toward 
the observer, and winding B shall have the 
perforations along the edge away from the 
observer. 

No preference for either type of winding is 
implied since both types are required for use 
on existing equipment. 

Film on Spools for Daylight Loading 

When the film is wound on a spool with a 
square hole in one flange and a round hole 



film with the perforations along one edge. 

With the types of winding described below, 
the emulsion side of the film shall face the 
center of the roll. 





Winding B 
Emulsion side in 



in the other flange, it shall be specified as 
winding B when wound as described for B 
above and with the square hole on the side 
away from the observer. 

Windings other than winding B, on spools, 
are considered as special-order products. 

Appendix 

(This Appendix is not a part of the Proposed 
American Standard for A and B Windings of 16mm 
Single-Perforated Film.) The types of winding cov- 
ered by this standard are limited to those which are 
in general use. 

It is recognized that film on spools, with a square 
hole in one flange and a round hole in the other, 
can be wound in other ways than that described as 
winding B, and that for special purposes these wind- 
ings may be supplied commercially. 



NOT APPROVED 



February 1953 Journal of the SMPTE VoL 60 



189 



SMPTE - Exhibitor Conference on 3-D 



Stereo, Cinerama, Cinemascope, wide- 
screen and stereophonic developments 
have added yeast to the motion picture 
cake and the dough is already rising. 
The practical application of these inno- 
vations raises serious questions of operat- 
ing standards both in production and 
exhibition. The major producers, 
through the Motion Picture Research 
Council, are actively engaged at present 
in efforts to answer the production ques- 
tions connected with 3-D. In an effort 
to clear the air for the exhibitors, the 
Society called and held a conference of 
the major exhibitor organizations (in- 
cluding two producer groups) on Feb- 
ruary 5, 1953. Invitations to attend the 
conference were sent to the organizations 
listed below, with but the latter three 
unable to send representatives: 

Allied States Association of Motion Picture 

Exhibitors Wilbur Snaper 
Metropolitan Motion Picture Theaters As- 
sociation Emanual Frisch 
Motion Picture Association of America 

John McCullough 

Radio City Music Hall Vincent Gilcher 
RKO Theatres Charles Horstman 
Theater Owners of America Samuel 

Pinanski 

Warner Brothers Frank Cahill 
United Paramount Theaters Harry Rubin 
Independent Theater Owners Association 

Harry Brandt 

Society of Independent Motion Picture Pro- 
ducers Ellis Arnall 

Representing the SMPTE were: Presi- 
dent, Herbert Barnett ; Engineering Vice- 
President, Henry Hood ; Chairman, Stereo- 
scopic Motion Pictures Committee, John 
A. Norling; Executive Secretary, Boyce 
Nemec; and Staff Engineer, Henry Kogel. 

Herbert Barnett chaired the meeting. 
At the outset, Mr. Barnett clearly out- 
lined the Society's position, "This meet- 
ing has been called by the SMPTE in 
an attempt to coordinate the engineering 
aspects of the development of systems 
utilizing third dimension and wide 
screen which have caught the imagina- 
tion of the American public in recent 



weeks." And to avoid any possible mis- 
understanding, he stated further, "It 
will be necessary to scrupulously avoid 
any attempts at comparative evaluations 
of competing systems, or the preference 
of one system or the individual features 
of one system over the others." 

To lay the groundwork for the discus- 
sion, John Norling sketched briefly the 
chief characteristics of the new develop- 
ments and the equipment changes, ex- 
hibitor-wise, required for each system. 

Cinerama: This system is designed to 
create the illusion of reality through 
panoramic effects. To do this three 
cameras are used in the taking process 
and three projectors are used in the 
theater. The three projected images are 
interlaced as a mozaic on a wide, curved 
screen to provide extreme wide-angle 
pictures involving the use of both front 
and peripheral vision. In addition, 6 
microphones are used for recording the 
sound from different action areas on 
separate sound tracks and 6 or more 
speakers are used in the theater to further 
aid the illusion of reality by having the 
sound come from the apparent source 
location. The sound tracks are recorded 
on a separate 35mm magnetic coated 
film. Installation of three projector 
booths, the special, wide curved screen, 
additional speakers and a magnetic 
sound reproducer are the required mini- 
mum equipment modifications. 

Cinemascope: Previously known as ana- 
morphoscope, this too is designed to cre- 
ate the illusion of reality through pano- 
ramic effects. Here special lenses are 
used on the camera and projector, the 
one to compress extreme wide-angle shots 
for standard 35mm negatives and the 
other to expand the compressed image 
for projection on a slightly curved screen 
roughly twice the present width. This re- 
quires, besides a wide screen and special 
lens, a higher projector light output. 



190 



3-D, Stereoscopic Motion Pictures: The 
design feature here is reality through the 
perception of depth. Present-day 3-D 
systems utilize two cameras, one for the 
left-eye image and the other for the right- 
eye image, simulating the process of 
normal vision. Two projectors, inter- 
locked to run synchronously, are then 
used with polarizing filters to project 
two separate images, one for the left eye 
and the other for the right eye. Viewing 
filters permit each eye to receive its own 
image as viewed originally by the two 
cameras. The brain fuses the two im- 
ages and the flat screen then provides 
images with apparent depth. 

Mr. Norling then listed the require- 
ments for 3-D showings : 

1. A metallic coated screen, 

2. Method of synchronizing the two pro- 
jectors and their shutters, 

3. Lenses matched for focal length and dis- 
tortion, 

4. Equal light output from each projector, 

5. Light output increased from each projec- 
tor by roughly a factor of 2, 

6. Larger magazines and reels for accep- 
tance of several intermissions. 

Mr. Barnett then asked how the Soci- 
ety could best serve the interests of the 
exhibitors. The reply to this boiled 
down to three essentials : 

1. Explain in laymen's language the new 
technical developments and future de- 
velopments as they occur. This would, 
in itself, eliminate much confusion in the 
trade. 

2. Supply unbiased answers to the ques- 
tions facing the exhibitors. 

3. Establish standards which will 

a. aidinterchangeability; 

b. retain as much of present equipment 
as possible; 

c. prevent newly purchased equipment 
from becoming obsolete; and 

d. permit readily interchangeable pro- 
jection of either 2-D or 3-D motion 
pictures. 

In regard to the first item. Mr. Nemec 
advised the group that the Society is in 
the process of preparing just such a story, 
which should be available in the near 
future. 

As for standards, he stated that the 
closest cooperation and liaison is being 



maintained between the SMPTE and 
the MPRC and that the above-men- 
tioned standards goals would undoubt- 
edly be used as a guide. 

The remainder of the conference was 
devoted primarily to expressing the most 
pressing questions before the exhibitors. 
These were listed as follows: 

Pertaining to 3-D 

On Screens 

\ . Can an all-purpose screen be developed 
for 2-D, 3-D and panorama pictures? 

2. Can existing screens, in good condition, 
be sprayed with a metallic paint to pro- 
vide a satisfactory surface for 3-D use? 

On Lenses 

3. What is the required accuracy of lens 
matching? 

On Filters 

4. Is there a nonfading type of polarizing 
projector filter? 

5. Has the projector filter been standard- 
ized? 

6. Has the viewing filter been standard- 
ized? 

7. Must the projector filters be artificially 
cooled? 

On Projection Lamps 

8. What portion of filters' light loss must 
be compensated for by increased lamp 
output? 

9. What light output differences between 
projectors can be tolerated? 

On Projectors 

10. What is the probable size of future 
magazines and reels? 

11. Are special arc supplies required for 
continuous projection? 

General Questions 

12. What are the chances of 3-D without 
viewing glasses or with a single pro- 
jector? 

13. What are splicing instructions for 3-D 
films? 

14. Are special rewinds and synchronizers 
needed? 

15. Explain apparent decrease in picture 
size. 

16. Explain in simple terms the funda- 
mental characteristics of the new de- 
velopments in motion pictures. 

17. Explain stereophonic and binaural 
sound. 

18. What can an exhibitor safely (to pre- 
clude rapid and expensive obsoles- 
cence) do now to prepare for scheduled 
3-D showings? 



191 



Pertaining to 2-D and Wide-Screen 
Processes 

19. Exhibitors recommended change of 
aspect ratio for 2-D pictures to give 
panorama effect with present equip- 
ment, shorter lenses and wider screens. 
What are possibilities of achieving this 
in the near future? 

The exhibitors were advised that these 
questions would be submitted to the 
Society's appropriate engineering com- 
mittees and the MPRC and that the 
answers would be forwarded to the ex- 
hibitors as soon as they became avail- 
able. (Several answers have since been 
made known and are listed at the end of 
this report.) 

At this point, the Society's committee 
structure was outlined and invitations 
were extended to the exhibitors to join 
and be represented through their tech- 
nically qualified people. 

In addition to asking questions, the 
exhibitors proposed that several of their 
recommendations be submitted to those 
concerned in the new developments : 

1. Any 3-D screen to be designed as an 
all-purpose screen capable of also be- 
ing used for normal and wide-screen 
motion pictures. 

2. Immediate consideration be given to 
increasing the present 4 : 3 aspect ratio 
to a picture width to height ratio 
more closely approximating 2:1. 
Here, it was felt that with a minimum 
of expense, panoramic projection 
could be immediately achieved to sup- 
ply renewed interest to large numbers 
of patrons. 

3. Immediate standardization of tech- 
nical terms to eliminate misunder- 
standing. Specific example was made 
of the term "3-D" which has been 
sorely misused. 

One of the questions posed previously 
was isolated far direct reply : What can an 
exhibitor safely do now to prepare for 
scheduled 3-D showings? Mr. Nemec 
stated that in his opinion 3-D is here to 
stay. The prospects are such as to insure 
relatively long term use of two-strip stereo 
equipment installed at this time. In con- 
nection with this, type of screen and reel 



size were discussed at some length. The 
reel size is dependent on the number of 
intermissions which are considered accept- 
able and upon mechanical limitations of 
present-day projection equipment. In a 
two- or three-projector theater, a minimum 
of one intermission will be unavoidable and 
a 24-in. magazine and 23-in. reel will cover 
all such cases. (The SMPTE was subse- 
quently informed that the MPRC had 
agreed on 25-in. magazines and 24-in. 
reels.) If two or more intermissions are 
acceptable then of course smaller maga- 
zines and reels can be used. 

Metallized screens will be required and 
will work equally well for all 3-D polarizing 
filter systems. They can also be used for 
regular 2-D projection, although a certain 
loss of brightness from side seats would 
result. 

Answers to several of the other questions 
listed have now become known: 

Q 2. Painting of existing screens may be 
satisfactory if carefully done so that per- 
forations are not filled. 

Q 5, 6. Polarizing filters for projection 
and viewing have been standardized. V- 
type filters, i.e., having planes of polariza- 
tion of 45 with the vertical, must be used. 

Q 10. In order to allow a show of 10,000 
ft with only one intermission, 25-in. maga- 
zines to accommodate 24-in. reels are 
recommended. These reels will hold up to 
5000 ft of color positive film or approxi- 
mately 5500 ft of black-and-white. Be- 
cause of their size and weight, such reels 
must have free-wheeling flanges to mini- 
mize strain on the film when the machine 
is started. Otherwise perforations will be 
pulled and the pictures will be out of 
synchronism. The MPRG believes that 
the spindle diameter may have to be in- 
creased. 

Q 13. The MPRC is preparing written 
instructions for theater projectionists which 
include splicing instructions. 

Q 14. The rewinds will have to be raised 
so that the larger reels will clear the rewind 
table, and the theaters will now require 
synchronizers. 

Q 18. Answered above. 

In addition to activities and reports 
forecast above, plans now are made to 
publish an excellent background and re- 
view article in the March Journal. It is 
John Norling's "The Stereoscopic Art," 
reprinted from the PSA Journal for 
November and December 1951, and 
January and February 1952. Henry 
Kogel, Staff Engineer. 



192 



73d Convention, Los Angeles Statler, April 27 - May 1 



The special features of this convention are 
increasing. Plans have previously been 
announced for giving special attention to 
outdoor theaters and for scheduling SMPTE 
Convention sessions to avoid conflict with 
ithe National Association of Radio and 
(Television Broadcasters' Seventh Annual 
j Conference which is at the Los Angeles 
Biltmore, April 28 - May 1 . 

Equipment Exhibit 

An equipment exhibit will be a feature 
of this Spring's SMPTE Convention. The 
hotel has made exhibit space available near 
the meeting rooms. A questionnaire and 
application blank are being sent to manu- 
facturers and distributors. Anybody wish- 
ing to exhibit new equipment or equipment 
to be discussed or demonstrated in papers 
at the Convention should write or wire 
Tom Gibbons, Minnesota Mining & Mfg. 
Co., 446 N. LaBrea Ave., Hollywood 36, 
Calif. 

Three-Dimensional 

Plans for consideration of three-dimen- 
sional, wide-screen and other current 
I developments may be : one or more papers 
will be presented to bring out or lead 
to discussion of particular problems or 
potentialities; or plans may materialize 
For an organized panel of speakers, with a 



moderator, to cover operational aspects 
such as: camera design and construction; 
camera operation in production shooting; 
laboratory problems in matching color 
balance, density, etc. ; projection problems 
such as interlocking, and large reel take-up ; 
stereophonic sound; and screen-brightness 
problems pertinent to stereoscopy. 

There may well be some reports on 
SMPTE standardization and engineering 
activities in the stereo and related fields. 

Reservations 

The Convention announcement card, 
giving the general plan of the sessions and 
the tear-off postal for hotel reservations is 
scheduled to be mailed on March 2. There 
are indications that hotel rooms may be 
in short supply, so those who can now plan 
their attendance should write for reserva- 
tions to: Mr. Thomas O'Hara, Front 
Office Manager, Hotel Statler, Wilshire 
Blvd. and Figueroa St., Los Angeles, Calif. 
Rates are: 

Single S 7.00 to $14.00 

Double 9. 50 to 14.00 

Twin 12. 00 to 16.00 

Parlor Suites 

(one bedroom) . . . 19.00 to 34.00 
Parlor Suites 

(two bedrooms) . . 21. 50 to 34.00 



Board of Governors Meeting 



At the January meeting of the Society's 
Board of Governors, there were the re- 
quired reports which are the record of 
what is past and there were the projected 
dollars and ideas budgets for 1953 and 
more. 

Reporting in person on the activities 
and responsibilities for their respective 
offices were: Herbert Barnett, John W. 
Servies, Henry J. Hood, Barton Kreuzer 
and William H. Offenhauser, Jr. Reports 
of the other officers were read for them. 
The complete roster of SMPTE officers 
and governors is given on the inside back 
cover of each Journal. 

Accomplishments of 1952 are already 
largely available to the membership in the 



form of the conventions and meetings held, 
the standards and engineering reports and 
the Journals published. The record of 
the business, financial and membership 
activities will appear again this year in the 
April Journal. 

A detailed operations report was given 
by Boyce Nemec. The relationships of 
types of income and expenditures were 
reviewed, with Gordon Chambers, a new 
Governor, analyzing the need for con- 
tinuing to chart graphically activities 
proposals and expenditures. The salient 
points from the operations report are 
reflected throughout the programs noted 
below. 



193 



Engineering 

Little will be said here : projects planned 
and in progress are reported regularly in 
the Journal by the Staff Engineer. Henry 
Hood presented to the Board drafts of four 
Standards which were approved for 
transmittal to ASA. These proposed 
standards have appeared in the Journal 
and will be published as final when ap- 
proved by the ASA. The advice and 
approval of the Board were sought re- 
garding enlarging the scope of the Screen 
Brightness Committee. The new scope 
will appear in the April Journal when the 
complete roster of SMPTE Committees 
is published. 

Charles Townsend outlined several as- 
pects of engineering related to television 
and film to which he suggested the Society 
should be certain to give attention. Axel 
Jensen clarified for the Board the duties 
and functions of the Joint Committee for 
Inter-Society Coordination which works 
to avoid overlapping of efforts among this 
Society, the Institute of Radio Engineers, 
Radio and Television Mfrs. Assn. and the 
National Assn. of Radio and Television 
Broadcasters. 

Publications 

Norwood Simmons' Editorial Vice- 
President's report was read, noting the 
excellent and vital work of Arthur C. 
Downes, Chairman, and the Board of 
Editors. Hard work by 1952's Convention 
Program Chairmen Geo. W. Colburn and 
Joe Aiken, with real help from the Papers 
Committee Vice-Chairmen, especially John 
Waddell for the International Symposium 
on High-Speed Photography, not only pro- 
duced well-attended sessions but also 
apparently reversed the trend toward a 
smaller volume of worthy papers for the 
Journal. Greater and more varied efforts 
by the Papers Committee are expected 
to produce an increasing volume of material 
which the membership wants. Those 
wants will be better known when the 
results of the Membership Service Question- 
naire are tabulated. 

On the basis of hopes for a reasonably 
greater volume of worthy Journal material 
for 1953, a budget was approved for 1328 
pp. 

194 



Conventions 

Jack Servies reported that he felt that 
the Spring Convention is in good hands 
and that all information from Hollywood 
thus far indicates a large, varied and 
successful convention. News of developing 
plans appear in each Journal's story about 
the^73d Convention. 

Sections 

A report for the Atlantic Coast Section 
was read by its new Chairman, William 
H. Offenhauser, Jr. Reports for the other 
sections were also read. An increased 
budget was put forth in view of the accom- 
plishments of 1952 and the plans for 1953, 
some of which are noted below in President 
Barnett's announced program. 

Geo. W. Colburn observed that the 
occasional reports of section meetings in 
the Journal had been very well received. 
It was agreed that a program of publishing 
the sections' quarterly reports, or pref- 
erably monthly meeting reports, would 
help keep the membership informed. 
Two such reports appear in this issue. 

New Plans 

So that the Society will be prepared, 
when it can afford such a program, the 
Board approved the appointment of a 
committee to examine the advisability of 
extending the Society's activities in order 
to promote more effectively the purposes 
for which this Society was founded, and 
particularly to examine the advisability 
of the Society's engaging in, promoting or 
financing scientific research, research fel- 
lowships and such other scientific projects 
directly promotive of the purposes of the 
Society. 

From the report by President Barnett, 
we have abstracted the introduction and 
the six points in the Society's program as it 
is presently expanded: 

The importance of long-range film and 
television engineering is exemplified by 
the film industry's thorough preparation 
for FCC theater television hearings that 
were renewed on January 26, and the list 
of well-qualified witnesses scheduled to 
appear, as an outgrowth of three earlier 
theater television channel appeals pre- 
sented before FCC by the Society. Ex- 
perimental channels secured by SMPTE 
gave studio and theater companies an 






opportunity to determine by actual practice 
the form that a national theater television 
service might take. Also, the current 
public interest stimulated by three- 
dimensional pictures, Cinerama, new types 
of screens and other dramatic innovations 
likely to appear are the result of many 
years of research and development. 

Improved public relations are needed 
by SMPTE to insure that engineers new 
to motion pictures and television are 
made aware of engineering services and 
standards information available from the 
Society. These six points are especially 
emphasized in the expanded program: 

1. Form new SMPTE subsections in 
cities where film and television engineers 
need help and stimulation from joint 
meetings. 

2. Offer counseling assistance to col- 
leges and universities interested in prepar- 
ing engineering students for careers in 
motion pictures and television. 

3. Find gaps in the published engineer- 
ing literature and offer assistance in filling 
them. 

4. Invite more active participation from 



other technical societies, trade associations 
and cultural groups in technical activities, 
and through these channels encourage edu- 
cational use of motion pictures, television 
and theater television. 

5. Publish special engineering studies 
for the benefit of businessmen, engineers 
and operating personnel, giving informa- 
tion on the functions, applications and 
effects of current technical developments. 

6. The Society's public relations activi- 
ties must emphasize (a) the need for special 
training of young engineers for work in 
motion pictures, (b) the need for improved 
technical quality in classroom motion 
pictures and in the manner of presentation, 

(c) the need for improved technical quality 
of films made for television to avoid a 
bad trade reaction that would adversely 
affect future markets for such films and 

(d) the need for television broadcasters, 
motion picture companies and theater 
circuits to be constantly on the lookout for 
new products and processes available 
commercially or through research and 
development programs which they wholly 
or partially support. V.A. 



Pacific Coast Section Meeting 



The first Pacific Coast Section meeting 
of 1953 was held on Tuesday evening, 
January 20, at the Filmcraft Television 
Theater, Hollywood, with an audience of 
approximately 400. The program featured 
"Rapid Drying of Normally Processed 
Black-and- White Motion Picture Films" 
by F. Dana Miller, Eastman Kodak Co., 
Rochester, N.Y., which was presented 
at the 72d Semiannual Convention at 
Washington and repeated here for the 
benefit of local members who had not 
attended the Convention. 

Dr. Karl Freund, President of Photo- 
Research Corp., read a paper describing 
a newly developed direct-reading photo- 
electric brightness meter. The meter 
was demonstrated and various expected 
applications of the instrument were men- 
tioned, including measurement of motion 
picture screen brightness, set illumination 
and kinescope brightness or contrast. 

In a progress paper on Eidophor, the 
new system for color theater television, 



Lorin D. Grignon discussed engineering 
developments in the Eidophor research 
program to date. Results of the first 
practical theater demonstrations of the 
Eidophor were described, as well as some 
of the problems which have been uncovered 
by these demonstrations. Members asked 
Mr. Grignon practical questions regarding 
future application of Eidophor and there 
was a very favorable general reaction, 
with enthusiasm for the technical 
knowledge that was made available. 

Membership Chairman J. W. Duvall 
distributed approximately 100 application 
blanks at the meeting. A show of hands 
revealed that 25% of those attending 
were not members. It is our hope that 
the new year will bring a sizable aggregate 
of new membership. 

The February meeting will feature a 
vital current topic stereo-photography. 
Philip G. Caldwell, Secretary-Treasurer, 
Pacific Coast Section. 



195 



Central Section Meeting 



The Section's Board of Managers met on 
January 29, before the papers' program 
at the Western Engineers Society, Chicago. 
As a delegate from the Section to the 
Illuminating Engineering Society, the 
Board chose Carl F. Jenson. Plans were 
made to continue the Section's present 
format of a letter-size meeting notice and 
to systematize the use of the lapel cards 
by the Section for membership promotion 
and program planning. In discussing the 
types of meeting reports, it was felt that 
a combination of business and management 
information and of the papers program 
might best serve the membership. An 
assessment is being made of the costs 
and values of sending notices of both 
section and subsection meetings to those 
in the Southwest. 

The Central Section heard William P. 
Kusack, Chief Engineer of television 
station WBKB, Chicago, talk about 
"Production Practices for Television" and 
Gordon Ray of Reid H. Ray Film In- 



dustries, Inc., St. Paul, discuss "Color 
Continuity and Reproduction." 

Mr. Kusack pointed to the necessity of 
operating within boundaries of technical 
characteristics while aspiring to good 
visual television reproduction. Basic rules 
need to be formulated, he said. The 
television transcription film "Television 
Lighting," produced by CBS, was shown 
to demonstrate the technical boundaries 
and supplement the paper. 

Mr. Ray discussed the building of color, 
design and continuity of various scenes 
in motion picture films and slides and he 
went into the problems of making photo- 
graphic color reproductions which simulate 
the original color film or photograph. He 
also covered the basic correlation between 
slide films and motion pictures to tell a 
story visually. 

The next Central Section meeting will 
be held on February 19. James L. Wassell, 
Secretary-Treasurer, Central Section. 



Magnetic Striping 



Plans and hopes now are to publish as 
Part II of the April Journal all seven of the 
papers on the Friday afternoon session, 
October 10, at the Washington Con- 
vention. Also, a paper which fits into this 
group has been made available by the 
Audio Engineering Society. The sym- 
posium was held under the chairmanship 
of Glenn Dimmick. Our efforts to publish 



the group, complete with discussions from 
the Washington Convention and from the 
Central Section Meeting, have been greatly 
enhanced by the extensive and careful 
work of R. T. Van Niman who has organ- 
ized the transcribing, editing, circulating 
and preparing of final corrected copy of 
the extensive discussion. The papers are 
listed below. V.A. 



"Manufacture of Magnetic Recording Materials" by Edward Schmidt and Ernest W. 

Franck, Reeves Soundcraft Corp. Presented on October 18, 1951, at the SMPTE 

Hollywood Convention. 
"Commercial Experiences With Magna-Stripe" by Edward Schmidt, Reeves Soundcraft 

Corp. Presented on October 10, 1952, at the SMPTE Washington Convention. 
"Magnetic Striping Techniques and Characteristics" by B. L. Kaspin, A. Roberts, H. 

Robbins and R. L. Powers, Bell & Howell Co. Presented on October 10, 1952, at 

the SMPTE Washington Convention. 
"Magnetic Striping of Photographic Film by the Laminating Process" by A. H. Persoon, 

Minnesota Mining and Manufacturing Co. Presented on October 10, 1952, at the 

SMPTE Washington Convention. 
"Magnetic Sound Tracks for Processed 16mm Motion Picture Film" by Thomas R. 

Dedell, Eastman Kodak Co. Presented on October 10, 1952, at the SMPTE 

Washington Convention. 
"Notes on Wear of Magnetic Heads" by G. A. Del Valle and L. W. Ferber, RCA Victor 



196 



Div., Camden, N.J. Presented on October 10, 1952, at the SMPTE Washington 

Convention. 
"A Study of Dropouts in Magnetic Film" by Ernest W. Franck, Reeves Soundcraft Corp. 

Presented on October 10, 1952, at the SMPTE Washington Convention. 
"Methods of Measuring Surface Induction of Magnetic Tape" by J. D. Bick, RCA Victor 

Div., Camden, N.J. Presented on October 29, 1952, at the 4th Annual Convention 

of the Audio Engineering Society. 
"Standardization Needs for 16mm Magnetic Sound" by E. W. D'Arcy, De Vry Corp. 

Presented on October 10, 1952, at the SMPTE Washington Convention. 



Current Literature 



The Editors present for convenient reference a list of articles dealing with subjects cognate to motion 
picture engineering published in a number of selected journals. Photostatic or microfilm copies of 
articles in magazines that are available may be obtained from The Library of Congress, Washington, 
D.C., or from the New York Public Library, New York, N.Y., at prevailing rates. 



American Cinematographer 

vol. 33, Oct. 1952 
Two New 16mm Films (p. 436) J. Van Natta 

vol. 33, Nov. 1952 

And Now . . . Cinerama (p. 480) J. W. Boyle 
Why I Used the Garutso Lens in Filming "The 

Four Poster" (p. 482) H. Mohr 
Variable Shutter for the Bolex H-16 (p. 484) 

F. Foster 
Carefully Balanced Lighting Vital to Best TV 

Film Results (p. 486) P. Tannura 

vol. 33, Dec. 1952 

The Development of Follow-Focus in Cine- 
matography (p. 523) F. Foster 
Overhead Lighting for Overall Set Illumination 

(p. 528) J. Ruttenberg 

Techniques for TV Commercials (p. 532) W. R. 
Witherell, Jr. 

American Scientist 

vol. 40, Oct. 1952 

Underwater Television and Marine Biology (p. 
679) H. Barnes 

Audio Engineering 

vol. 36, Oct. 1952 

Handbook of Sound Reproduction, Chapter 5 : 
"Musical Instruments and the Human Voice" 
(p. 36) E. M. Villchur 

vol. 36, Nov. 1952 

New Medium-Cost Amplifier of Unusual Per- 
formance (p. 30) G. L. Werner and H. Berlin 
Handbook of Sound Reproduction (p. 40) E. M. 
Villchur 

vol. 36, Dec. 1952 

Handbook of Sound Reproduction (p. 20) E. M. 
Villchur 

Bild und Ton 

vol. 5, Nov. 1952 
Die Lebensdauer der Filmkopien in Abhangig- 

keit von den Abmessungen der Transportrollen 

(p. 347) K. 0. Frielinghaus 



Kinotechnische Normung : 
ISO-Tagung (p. 356) 



Bericht iiber die 



British Kinematography 

vol. 21, Aug. 195: 
High-Definition Films (p. 32) N. Collins and 

T. C. Macnamara 
A New Television Recording Camera (p. 39) 

W. D. Kemp 

vol. 21, Oct. 1952 
International Conference on Cinematographic 

Standards (p. 88) 
Modern Kinema Lighting (p. 93) G. Robinson 

vol. 21, Nov. 1952 
Considerations Affecting the Design of Television 

Cameras (p. 117) G. C. Newton 
Inlay Process for Television Production (p. 122) 

A. M. Spooner 
Use of the Radio Talk-Back Unit in Television 

Productions (p. 127) R. Toombs 

Electronics 

vol. 25, Nov. 1952 

Evaluating AFC Systems for Television Receivers 
(p. 132) G. Howitt 

International Photographer 

vol. 24, Nov. 1952 
Shooting Live TV Shows with Motion Picture 

Cameras (p. 5) K. Freund 
Television Filming (p. 8) V. E. Hughes 

vol. 24, Dec. 1952 

The Natural Vision 3D Camera (p. 5) 
Cinerama (p. 9) A. Nadell 

International Projectionist 

vol. 27, Oct. 1952 
Safety Film: Performance Characteristics (p. 5) 

R. A. Mitchell 
Cinerama A Step in the Right Direction (p. 10) 

A. Nadell 



197 



vol. 27, Nov. 1952 
Safety Film: Performance Characteristics, Pts. 

II, (p. 6) R. A. Mitchell 
Transistor Successor to the Vacuum Tube? 

(p. 12) 

De Vry's JAN Magneto-Optical Portable Pro- 
jector (p. 19) 

vol. 27, Dec. 1952 
Projector Mechanisms Have Improved (p. 7) 

L. Chadbourne % 

The New Ansco Color Film and Process (p. 13) 

R. A. Mitchell 

Kinematograph Weekly (Ideal Kinema section) 

vol. 18, Oct. 9, 1952 

Optics of the Sound Head (p. 13) R. H. Cricks 
Cricks Sees Synchro-Screen (p. 11) R. H. Cricks 

Kino-Technik 

vol. 6, Oct. 1952 

Tonschriften im Gegentaktverfahren (p. 232) 
Filmaufzeichnugsanlage im Fernsehbetrieb (p. 

236) 
Britisches Farbfernsehen vor deutschen Augen 

(p. 237) 
Storungen bei der Vorfuhrung von Tonfilmen 

(p. 245) K. Braune and H. T'mmel 

no. 11, Nov. 1952 
Fernsehkameras ferngesteuert und fernbetrie- 

ben (p. 252) 
Kinematische Fragen an Filmschaltgetrieben (p. 

262) H. Weise 

vol. 6, Dec. 1952 
Zum Normblatt-Entwurf Sicherheitsfilm (p. 276) 

L. Busch 
Verwendung von HI-Kohlen in der Schmalfilm- 

projektion (p. 279) C. Heimann 
1 200-m-Schmalfilmspuler ermoglichen pausen 

lose Vorfuhrung (p. 280) E. May 
Die Fernseh-Grossprojektion im Kino (p. 282) 

F. Winckel 
Storungen bei der Vorfuhrung von Tonfilmen 

(p. 290) K. Braune and H. Tummel 



Motion Picture Herald 

vol. 189, Nov. 22, 1952 

Natural Vision Ready for Public Showing (p. 38) 
W. R. Weaver 

Motion Picture Herald 

vol. 189, Dec. 6, 1952 
(Better Theaters Section) 

What Projectionists Should Know About Film 
Stock Today (p. 37) G. Gagliardi 

Philips Technical Review 

vol. 14, July 1952 

High-tension Generators for Large-Picture Pro- 
jection Television (p. 21) J. J. P. Valeton 

Photographic Journal 

vol. 92B, Sept.-Oct. 1952 
(High-Speed Photography Issue) 
Flash Cinematography (p. 129) R, H. J. Brown 
A Synchronized Flash-Discharge System for 

High-Speed 35mm Cinematography (p. 133) 

W. D. Chesterman and G. T. Peck 
Image Converter Tubes and Their Application 

to High-Speed Photography (p. 137) J. S. 

Courtney-Pratt 

Image Converter Techniques Applied to High- 
Speed Photography (p. 149) R. A. Chippendale 
An Electronically Operated Kerr Cell Shutter 

(p. 158) K. D. Froome 
Electro-Optical Shutters as Applied to the Study 

of Electrical Discharges (p. 161) J. M. Meek 

and R. C. Turnock 

Radio & Television News 

vol. 48, Nov. 1952 

Miniature TVI Wavetraps (p. 41) R. P. Turner 
Television "Snow" (p. 58) W. H. Buxhsbaum 
Self-Focus Picture Tubes (p. 122) E. M. Noll 

Radio & Television News 

vol. 48, Nov. 1952 

(Radio-Electronic Engineering Section) 
Audio Facilities for TV Studios (p. 3) E. P. 

Vincent 
Color TV Definitions (p. 32) 



Book Review 



Exposure Meters and 
Practical Exposure Control 

By J. F. Dunn. Published (1952) by The 
Fountain Press, 46-47 Chancery Lane, 
London WC2, England. 252 pp. (incl. 10 
pp. index) + 8 pp. adv. Numerous tables; 
97 illus. and plates. 6^ X 8 in. Price 
35 shillings. 

Technical aids to the control of photo- 
graphic exposure have always been a mat- 
ter of lively interest to photographers of all 



types and have occasionally been the sub- 
ject of passionate and perhaps excessively 
partisan debate. As long as still mono- 
chrome photography with wide-latitude 
materials was all that was involved, excel- 
lent results could be obtained with a moder- 
ately wide range of exposures, and the dis- 
crepancies between the various devices 
employed and the techniques used were 
not of great importance. In recent years 
however, the accuracy demanded of ex- 
posure estimating equipment for color 



198 



[photography, for high-grade motion pic- 
ture work, and for many specialized photo- 
graphic processes has sharpened the debate 
and brought into much clearer focus the 
relation between the light distribution of the 
original scene and that of the final photo- 
graphic product. 

Mr. Dunn's book makes available to the 
working photographer, amateur or pro- 
fessional, a single source describing virtu- 
ally all the equipment and techniques for 
determining exposure. A chapter on the 
fundamental theory of exposure require- 
ments presents an informative discussion 
of the derivation of film speed ratings, the 
criteria of correct exposure, and the meth- 
ods of assessing the subject lighting, includ- 
ing brightness measurements (highlight, 
shadow, average, and "keytone") and 
"incident light" evaluation. Chapters on 
exposure tables and calculators, extinction 
meters, photoelectric integrating meters, 
photoelectric incident light meters and 
exposure photometers include detailed de- 
scriptions of most available commercial 
equipment (European and American). 
Useful intercomparisons of the results ob- 
tainable and the uses, limitations, and pre- 
cautions to be observed in each case are 
provided. Throughout the book, the dif- 
fering requirements of still monochrome 
photography and motion picture and color 
work are emphasized. 

Many tables are included, such as film 
speed ratings, comparisons of various speed 
rating systems, etc. Most of the working 
data that involves speed ratings is pre- 
sented in a double notation giving the B.S. 
(British Standards Institution) and A.S.A. 
ratings side by side. The exposure tables 
however are presented for use with the 
B.S. ratings only. American readers will 
prefer equivalent tables based on A.S.A. 
ratings, such as the A.S.A. Photographic 
Exposure Computer for daylight photog- 
raphy, but may find it worth while to con- 
vert to B.S. ratings in order to make use of 
the excellent tables for photography by 
artificial light. 

Only a very few minor criticisms might 
be made of Mr. Dunn's book, and these do 



not detract from the general excellence of 
the presentation. The author has gone too 
far in an effort to substitute nontechnical 
language for relatively simple technical 
concepts. I suspect for example that the 
use of the term "kissing" for "tangent" 
would be more likely to involve the reader 
in speculations as to the aptness of the 
metaphor than it would be likely to clarify 
the concept of tangency. In discussing the 
variation of light output of incandescent 
lamps with voltage (in a table on page 83 
and in connection with the calibration of 
exposure meters on page 129) the actual 
voltage variations are used rather than the 
percentage variation. Inasmuch as the 
lamps described are operated at about 230 
volts, the voltage figures as given are not 
valid for lamps operated at 115 volts or 
lower. If the variation had been given in 
per cent, the data would be sufficiently 
accurate for lamps operated at any voltage. 
Although exposure tables and calculating 
devices are described extensively, tables 
of "guide numbers" commonly used at 
least in the United States with photoflood 
and photoflash light sources are not men- 
tioned. 

Mr. Dunn devotes considerable space to 
a description of exposure photometers, 
particularly the SEI meter, for which he is 
so largely responsible. It should be noted 
that although his perhaps pardonable en- 
thusiasm for this instrument colors the 
parts of the book in which its construction 
and use are described, it has not affected 
the sections dealing with other types of 
instruments, which are described fully and 
fairly. 

Whether the reader is already committed 
to using a particular kind of exposure meter, 
or wishes to determine what meter to ob- 
tain or what techniques to use, or even if 
he is not interested in a meter at all and 
wants only a general understanding of the 
problem together with a set of useful 
tables, he will find the book instructive 
and helpful. Theodore H. Projector, Na- 
tional Bureau of Standards, Washington 25, 
D.C. 



A new edition of the Society's Test Film Catalog is now available at no charge from the 
Society's headquarters. It covers 27 different test films, 16mm and 35mm, for use by 
theaters, service shops, factories and television stations. These test films have been 
developed by the SMPTE and the Motion Picture Research Council. 



199 



New Members 



The following members have been added to the Society's rolls since those last published. The 
designations of grades are the same as those used in the 1952 MEMBERSHIP DIRECTORY. 



Honorary (H) 



Fellow (F) 



Active (M) 



Associate (A) 



Student (S) 



Adams, John H., Technical Chief Radio and 
Television Engineer, KFDX, KFDM. Mail: 
2207 Grant St., Wichita Falls, Tex. (M) 

Anderson, Louis L., Manager, Magnetic 
Equipment Dept., Brush Development Co. 
Mail: 3405 Perkins Ave., Cleveland 22, 
Ohio. (M) 

Baker, Stanley E., Photographer, North Ameri- 
can Aviation. Mail: 8321 Keokuk Ave., 
Canoga Park, Calif. (A) 

Baumert, Ernest, Supervisor, Technical Main- 
tenance Branch, Signal Corps Pictorial Center, 
51-01 39 Ave., Long Island City 4, N.Y. 
(M) 

Berti, Nullo, Salesman, 56-01 137 St., Flush- 
ing 55, N.Y. (A) 

Bibas, Frank P., Motion Picture Director, 
McCann-Erickson, Inc. Mail: Blindbrook 
Lodge, Rye, N.Y. (M) 

Blair, E. M., Chief Engineer, E.D.L. Co. Mail: 
1240 Clay St., Gary, Ind. (M) 

Brown, Sam E., Assistant Executive Director, 
Academy of Motion Picture Arts and Sciences. 
Mail: 177 South Citrus Ave., Los Angeles 36, 
Calif. (M) 

Bryant, William E., Television Engineer, Na- 
tional Broadcasting Co., Sunset & Vine, 
Hollywood 28, Calif. (A) 

Burtis, Eric F., Director, Post Film Library 
Services, U.S. Army Signal Corps, Signal 
Office, Ft. Eustis, Va. (A) 

Gaboon, Roy D., Prairie Regional Engineer, 
Canadian Broadcasting Corp. Mail: 143 
Wildwood Park, Winnipeg, Manitoba, 
Canada. (M) 

Cane, Albert Chemical Engineer, Technicolor 
Motion Picture Corp. Mail: 856 South 
Normandie Ave., Los Angeles 5, Calif. (A) 

Cole, Lionel J., Film Producer, Shell Caribbean 
Petroleum Co., Apartado 809, Caracas, 
Venezuela. (M) 

Connelly, James J., Project Engineer, Sperry 
Gyroscope Co. Mail: 2327 Andrews Ave., 
New York 68, N.Y. (A) 

Gopeland, William H., Radio-Television-Broad- 
cast Engineer, C.B.S. Television. Mail: 255 
South Amalfi Dr., Santa Monica, Calif. (M) 

Curtis, Sgt. Charles F., Motion Picture Tech- 
nician, 3206 Photo Test Squadron, Eglin Air 
Force Base, Fla. (A) 

De Rycke, Lawrence F., Illustrative Photog- 
rapher, GMC Truck & Coach Div. Mail: 
27 North Baldwin Rd., R #1, Lake Orion, 
Mich. (A) 

Dieffenbach, Capt. Woods W., AO-577368, 
Officer in Charge of Motion Picture Section, 



32,06 Photographic Test Squadron, Eglin Air 

Force Base, Fla. (M) 
Diehl, Adam E., Dean, Los Angeles City College. 

Mail: 5056 Ambrose, Los Angeles 27. (A) 
Donneau, Peter J., Television Engineer, 

WJAR-TV. Mail: 26 Monty St., Woon- 

socket, R.I. (A) 
Fierst, Morris E., General Manager, Packaged 

Programs, Inc., 634 Penn Ave., Pittsburgh 22, 

Pa. (A) 
Fisher, Dennis R., Technical Manager, Kodak 

South Africa, P.O. Box 735, Capetown, South 

Africa. (A) 

Ford, K. A., Partner, Triangle Continuous Day- 
light Motion Picture Projector Co. Mail: 

537 Woodlawn Ave., Glencoe, 111. (M) 
Frothingham, Anthony, Sales Engineer, Kodak- 

Pathe' S.A.F., 17 rue Francois ler., Paris 

8 erne., France. (M) 

Galbreath, Richard E., Motion Picture Pro- 
ducer, Galbreath Picture Productions, Inc., 

2905 Fairfield Ave., Ft. Wayne, Ind. (M) 
Goldberg, Benedict S., Jr., Motion Picture 

Cameraman, Lockheed Aircraft Corp. Mail: 

6927 Aura Ave., Reseda, Calif. (A) 
Gray, David Anton, Williams College. Mail: 

Box 564, Williamstown, Mass. (S) 
Gustafson, G. E., Vice-President in Charge of 

Engineering, Zenith Radio Corp., 6001 W. 

Dickens Ave., Chicago, 111. (M) 
Hales, Frederick J., Recordist, Warner Brothers 

(England). Mail: 2 Cordigan Mansions, 

Richmond Hill, Richmond, Surrey, England. 

(A) 
Hanna, Clifford, Motion Picture Producer, 

Video Films, 1004 East Jefferson Ave., 

Detroit, Mich. (A) 
Hilfinger, Harry P., ESO-S Pictures. Mail: 

828 W. 39 St., Kansas City, Mo. (M) 
Hirschkop, Morton A., New York University. 

Mail: 255 South Third St., Brooklyn 11, 

N.Y. (S) 
Hulcher, Charles A., President, Charles A. 

Hulcher Co., Inc. Mail: 40 Manteo Ave., 

Hampton, Va. (A) 
Jaime, Jose' de Lugo, Laboratory Assistant, 

Especialidades Filmicas. Mail: Empedrado 

#360, Apto. 311, Havana, Cuba. (A) 
Jelsma, Charles E., Motion Picture Technician, 

Berndt-Bach. Mail: 1880 Riverside Dr., 

Los Angeles 39, Calif. (A) 
Karson, Walter E., Motion Picture Equipment 

Rebuilding, Edwar3 H. Wolk. Mail: 5750 

South Elizabeth St., Chicago 36, 111. (A) 
Kurtz, Jerome, New York University. Mail; 

3150 Rochambeau Ave., New York (S) 



200 



Levine, Harold H., Sound Dubbing, Ryder 
Service, Inc. Mail: 1809 Pass Ave., Burbank, 
Calif. (A) 
Lidner, Scott M., New York University. Mail: 

1305 President St., Brooklyn 13, N.Y. (S) 
Lipman, Robert N., Mechanical Design Engi- 
neer, RCA Victor Div. Mail: 235 Lawnside 
Ave., Collingsvvood 7, N.J. (A) 
Luce, R. Robert, Supervising Editor, Geo. W. 
Colburn Laboratory. Mail: 1954 Farwell 
Ave., Chicago 26, 111. (A) 

Lukas, Walter, Engineer, Emerson Radio & 
Phonograph Corp. Mail: 115 Belvidere 
Rd., Glen Rock, N.J. (A) 
Martin, Glenn C., Jr., Stage Lighting Designer 
and Assistant Manager, Texas Scenic Co. 
Mail: 1255 Fulton Ave., San Antonio 1, 
Tex. (A) 

Mian, Attilio, Recording Engineer, Fulton Re- 
cording Co. Mail: 1686 Grand Concourse, 
New York, N.Y. (A) 
Micco, Leopoldo A., Physicist, Ansco. Mail: 

125 Leroy St., Binghamton, N.Y. (M) 
Morgan, Miles, Photographer, U.S. Army. 
Mail: 6350 Franklin Ave., Hollywood. (A) 
Nash, John S., Motion Picture Photographe 
and Editor, 7313 Santa Monica Blvd., Holly- 
wood 46, Calif. (M) 
Nortman, Richard P., Filming Director, Arizona 

State College, Tempe, Ariz. (A) 
Pelletier, Claude, Recordist, National Film 
Board of Canada, John St., Ottawa, Ontario, 
Canada. (A) 

Pfening, Fred D., Jr., The Fred D. Pfening 
Co., 1075 West Fifth Ave., Columbus, Ohio. 
(A) 

Philippe, Lauwers, Chief Chemist, N. V. 
Gevaert Photo-Producten. Mail: 21 L. 
Gerritz Ave., Berchem, Antwerp, Belgium. 
(A) 

Raymond, Julian E., Audio Engineer, Assistant 
Cameraman, Set Designer, TV Ads, Inc., 
3839 Wilshire Blvd., Los Angeles, Calif. (M) 
Rett, Hubert C., Television Engineer, National 
Broadcasting Co. Mail: 14 West Elm St., 
Chicago 10, 111. (A) 

Rosenberg, Irving, Technical Supervisor, 
Columbia Broadcasting System. Mail: 98- 
34 63 Dr., Forest Hills, N.Y. (A) 
Rosenberg, Jerome M., Electrical Engineer, 
Chromatic Television Laboratories, Inc., 
703 37 Ave., Oakland 1, Calif. (A) 
Ruppert, Clyde R., Motion Picture Film Editor, 
Geo. W. Colburn Laboratory. Mail: 4027 
North Maplewood Ave., Chicago 18,^ 111. (A) 
Sargente, Mario, Toolmaker, Pathe' Labora- 
tories, Inc. Mail: 2339 Prospect Ave., 
Bronx, N.Y. (A) 

Saunders, Bernard G., Physicist, Oak Ridge 
National Laboratory. Mail: 100 Plymouth 
Cir., Oak Ridge, Tenn. (A) 
Scripps, William J., Telecommunications Con- 
sultant, W. J. Scripps Associates, Inc., 286 
S. Woodward, Birmingham, Mich, (M) 



Seyfried, Grover, Director of Photography, 
Soundfilm Studios, Inc. Mail: 4815 Cabot 
Ave., Detroit, Mich. (M) 

Shapiro, Irvin, Motion Picture and Television 
Executive, Standard Television Corp. Mail: 
565 Park Ave., New York 21, N.Y. (M) 
Shibuk, Charles, New York University. Mail: 
2084 Bronx Park East, New York, N.Y. (S) 
Singleton, Harold C., Consulting Radio Engi- 
neer, Chief Engineer, KGW, KGW-TV. 
Mail: 4488 SW Council Crest Dr., Portland 
1, Ore. (M) 

Stewart, Henry H., Motion Picture Photog- 
rapher, Bureau of Public Roads, Dept. of 
Commerce. Mail: 147 Fleetwood Ter., 
Silver Spring, Md. (A) 

Taylor, Frank Nash, Works Manager, Kodak 
South Africa, 102 Davies St., Doornfontein, 
Johannesburg, South Africa. (A) 
Theis, H. Grant, Manager, Film Service Opera- 
tions Dept., CBS Television. Mail: 440 East 
Palisade Ave., Englewood, N.J. (M) 
Umbarger, Ralph, Cameraman, Rarig Motion 
Picture Co. Mail: 302 19 Ave. South, 
Seattle 44, Wash. (A) 

Watterlohn, R. H., Electronics Engineer, Bell 
& Howell Co. Mail: 5448 West Huron 
St., Chicago 44, 111. (M) 

Weber, Carlton F., Director of Photography and 
Television Recording, U.S. Air Force. Mail: 
1110 North Mariposa, Burbank, Calif. (A) 
Westfall, Robert M., Officer, U.S. Navy. Mail: 

503-B Saratoga, China Lake, Calif. (M) 
Williams, Charles J., Cinetechnician, Sub- 
foreman, Unicorn Theaters, Inc. Mail: 
14825 Fox St., San Fernando, Calif. (A) 
Wood, Lt. Douglas R., Motion Picture Officer, 
U.S. Air Force, 3206 Photo Test Squadron, 
Eglin Air Force Base, Fla. (A) 

CHANGES IN GRADE 

Auerbach, Gerald, (S) to (A) 
Greenberg, Raymond, (S) to (A) 
Jeffery, Seymour, (A) to (M) 
Little, Ralph V., Jr., (A) to (M) 
Pearson, Lloyd K., (A) to (M) 
Thome, Frederick R., (A) to (M) 
Ungar, Albert J., (S) to (A) 

DECEASED 

Chatelain, Arthur B., Foreman, Laboratory, 
Twentieth Century-Fox Film Corp. Mail: 
701 S. Verdugo Rd., Apt. 3, Glendale 5, 
Calif. (A) 

Foss, William L., President, William L. Foss, 
Inc., 927-15 St., N.W., Washington, D.C. 
(M) 

Jones, Merwin C., Maintenance Supervisor, 
KGO-TV, American Broadcasting Co. Mail: 
270 El Bonito Way, Millbrae, Calif. (A) 

O'Toole, Russel, Sound Engineer, RCA Service 
Co. Mail: 1321 Spear, Logansport, I ml. 
(A) 

201 



Journals Available and Wanted 



These notices are published as a service to expedite disposal and acquisition of out-of-print Journals 
Please write direct to the persons and addresses listed. 

Available 

1951-1952 Journals in excellent condition plus the Indexes for 1916-30, 1930-35, 1936-45 
and 1946-50; and including the 1949 High-Speed Photography. For best offer write 
to K. C. Tsien, 147-51 Charter Road, Jamaica 35, N.Y. 

Wanted 

Transactions 1, 6 and 7. Write Mrs. Dorothy Gelatt, Henry M. Lester, 101 Park Ave., 
New York 17, N.Y. 

January and February 1946 Journals. Advise the Record Engineering Library, Radio 
Corporation of America, RCA Victor Division, 501 N. LaSalle St., Indianapolis, Ind. 



Meetings 



Society of Motion Picture and Television Engineers, Southwest Subsection Meeting, 

Mar. 16, Fort Worth, Tex. 

Inter-Society Color Council, Annual Meeting, Mar. 18, Hotel Statler, New York, N.Y. 
American Institute of Electrical Engineers, New York Section, Meeting on "High-Energy 
Accelerators," Mar. 19. Engineering Societies Bldg., New York 
Optical Society of America, Mar. 19-21, Hotel Statler, New York, N.Y. 
American Physical Society, Joint Meeting with APS Southeastern Section, Mar. 26-28. 

Duke University, Durham, N.C. 

Symposium on Modern Network Synthesis, planned by Polytechnic Institute of Brooklyn, 
Apr. 16-18, Auditorium of Engineering Societies Bldg., New York 

International Symposium on Nonlinear Circuit Analysis, Apr. 23-24, information from 
Microwave Research Inst., 55 Johnson St., Brooklyn 1, N.Y. 

73d Semiannual Convention of the SMPTE, Apr. 27-May 1, Hotel Statler, Los Angeles 
National Association of Radio and Television Broadcasters, 7th Annual Conf., Apr. 28- 

May 1, Ambassador Hotel, Los Angeles 

American Physical Society, Apr. 30-May 2, Washington, D.C. 
Acoustical Society of America, May 7-9, Hotel Warwick, Philadelphia, Pa. 
Society of Motion Picture and Television Engineers, Southwest Subsection, May 20, 

Dallas, Tex. 

Society of Photographic Engineers, Third Annual Conference on Science in Photography 
and Photographic Instrumentation, May 20-22, U.S. Hotel Thayer, West Point, N.Y. 
American Physical Society, June 18-20, Rochester, N.Y. 

American Institute of Electrical Engineers, Summer General Meeting, June 29- July 3, 

Atlantic City, N.J. 

Biological Photographic Association, 23d Annual Meeting, Aug. 31-Sept. 3, Hotel Statler, 

Los Angeles, Calif. 

The Royal Photographic Society's Centenary, International Conference on the Science 
and Applications of Photography, Sept. 19-25, London, England 

National Electronics Conference, 9th Annual Conference, Sept. 28-30, Hotel Sherman, 

Chicago 

74th Semiannual Convention of the SMPTE, Oct. 4-9, Hotel Statler, New York 
Audio Engineering Society, Fifth Annual Convention, Oct. 14-17, Hotel New Yorker, 

New York, N.Y. 

Theatre Equipment and Supply Manufacturers' Association Convention (in conjunction 
with Theatre Equipment Dealers' Association and Theatre Owners of America), 

Oct. 31-Nov. 4, Conrad Hilton Hotel, Chicago, 111. 

202 



Theatre Owners of America, Annual Convention and Trade Show, Nov. 1-5, Chicago, 111 
National Electrical Manufacturers Association, Nov. 9-12, Haddon Hall Hotel, Atlantic 

City, N.J. 

75th Semiannual Convention of the SMPTE, May 3-7, 1954 (next year), Hotel Statler, 

Washington, D.C. 

76th Semiannual Convention of the SMPTE, Oct. 18-20, 1954 (next year), Ambassador 

Hotel, Los Angeles 



Employment Service 



Position Available 

Permanent position in Southwest for 
experienced motion picture cameraman; 
must have sample interior and exterior 
footage to indicate ability. Write letter, 
giving re'sume' of professional experience, 
to Susong Agency, 524 Commercial Bldg. 
Dallas, Tex. All replies confidential. 



Positions Wanted 

Audio-Visual School of Education Gradu- 
ate: M.A., Audio-Visual Education, New 
York University. Sound background in 
personnel and contact work, attractive, 
single, personable. Prefer position New 
York or New Jersey area. Spent 3 years 
abroad, civilian, Special Services Director. 
Miss Fredericka Appleby, 810 Broadway, 
Newark, N.J. HUmboldt 5-4582. 



TV Producer-Director: Formerly Chief 
of Production in Army's first mobile TV 
system, experience in writing-directing 
high-speed, low-cost instructional pro- 
ductions; TV producer-director, KRON- 
TV San Francisco, five shows weekly. 
Desire connection in educational TV, 
preferably employing kinescope technique; 
married; prefer West Coast, but willing 
to travel; resume', script samples, pictures 
of work on request. Robert Lownsbery, 
1116 E. Claremont St., Pasadena 6, Calif. 



Resigning Feb. 1 as gen. mgr., charge 
of production, large southern film studio. 
15 yrs. experience as prod, mgr., editor 
and cameraman, 16mm and 35mm. 
Married, 37, college grad. References 
and resume on request Harlan H. 
Mendenhall, 1609 Blodgett, Houston 4, 
Tex. 



Motion pictures in color depend on the engineers' knowledge of the "Principles of 
Color Sensitometry." A 72-page article bearing that title and prepared by the Color 
Sensitometry Committee appeared in the Journal for June 1950. Attractive reprint 
copies may be purchased for $1.00. 



SMPTE Lapel Pins are available from 
the Society's headquarters. They are 
gold and blue enamel, with a screw back. 
The pin is a -in. reproduction of the 
Society symbol the film, sprocket and 



television tube which appears on the 
Journal cover. The price of the pin is 
$4.00, including Federal Tax; in New 
York City, add 3% sales tax. 



SMPTE Officers and Committees: A new publishing of the roster of Society 
Officers and the Committee Chairmen and Members is scheduled for the April 
Journal. The last one is in April 1952. 



203 



New Products 



Further information about these items can be obtained direct from the addresses given. 
As in the case of technical papers, the Society is not responsible for manufacturers' state- 
ments, and publication of these items does not constitute endorsement of the products. 





The Robot II Splicer is shown in the 
currently available Mark IV 35mm model. 
Developed in England by Gustav Jirouch, 
it is marketed in England by Cine Tele- 
vision Equipment Ltd., and is distributed 
in the United States by Hollywood Film 
Co., 5446 Carlton Way, Hollywood 27, 
Calif. Designed to be a very fast and fully 
automatic splicer, its splice width is given 
as 0.077 in. Its dimensions are 7J X 



8f X 6^ in. and it weighs 38 Ib. The price 
is $650.00, delivered. Particular ad- 
vantages claimed for it are : equal scraping 
of emulsion down to the base and removing 
the emulsion in one piece, thus minimizing 
dirt and dust; straight and square end 
cutting; elimination of any special sound- 
track blooping; and making double or 
treble joins without loss of material. 



204 



Sound-on-Film Recording Using 
Electrooptic Crystal Techniques 



By ROBERT DRESSLER and ALBERT A. CHESNES 



This paper deals with the theoretical and practical aspects of a sound-on-film 
recorder with no moving parts, utilizing the birefringence properties of 
certain crystals. The physical properties of various crystals, as well as final 
performance measurements of the entire sound system, are discussed. 



.T OR THE PAST ten years, Paramount 
Pictures has been engaged in the 
development and production of a theater 
television and video recording inter-film 
system. In a system of this type, both 
the sound and picture must be impressed 
on the film simultaneously. Conven- 
tional modulators used for sound record- 
ing are expensive, quite fragile and can 
easily be rendered inoperative by over- 
voltage. Since Paramount's inter-film 
equipment is designed to be operated by 
lay people for the most part, there came 
the need for a low-cost, rugged sound 
modulator. 

The light modulator to be described 
makes use of the physical properties of 
some materials which allow them to re- 
tard polarized light. Before beginning 
a detailed description of the modulator, 
it is well to become familiar with the 

Presented on October 10, 1952, at the 
Society's Convention at Washington, D.C., 
by Robert Dressier, who read the paper, 
and Albert A. Chesnes, Paramount Pictures 
Corp., 1501 Broadway, New York 36, 
N.Y. 



physical phenomena which make light 
modulation possible. 

General Considerations 

If a light source is placed behind 
crossed polarizers, essentially all light 
output from this assembly is obscured. 
Light dips of 300:1 or better are ob- 
tained with reasonable polaroids. It 
has been noticed that a large number of 
common solid transparent materials, 
when placed between these polarizers, 
will produce two rays travelling with 
different velocity so that the light be- 
comes elliptically polarized and produces 
a component of energy that will pass 
through the second polarizer. Cello- 
phane, lucite and polystyrene are just a 
few of the materials which will produce 
optical birefringence. There is also a 
class of liquids which when placed in a 
strong electric field will produce rotation of 
a polarized beam of light. This effect for 
liquids was discovered by Kerr and bears 
his name, the Kerr effect. 1 Kerr cells 
made with these liquids can and have 
been used as light modulators. How- 
ever, several difficulties appear when 



March 1953 Journal of the SMPTE Vol. 60 



205 




Fig. 1. Typical ADP crystal. 

using cells of this type. One important 
difficulty deals with liquid purity and 
manifests itself because many of the ma- 
terials which exhibit a high Kerr effect 
are deteriorated by the electric field re- 
quired to produce the effect. It there- 
fore becomes necessary to constantly 
pump new liquid through the cell to keep 
it pure, an obvious disadvantage. 

Continuing with materials which ex- 
hibit optical retardation when placed in 
an electric field, we come to a group of 
solids studied for the most part by 
Pockels, and the effect is known as the 
Pockels effect. 2 Examples of some of 
these materials are quartz, sodium chlo- 
ride and zinc sulfide crystals. These ma- 
terials are stable solids and form a suit- 
able group on which light modulation 
studies can be built. Almost every early 
experimenter who has worked with the 
Pockels effect cites the zinc sulfide crys- 
tal as the one with the ideal property 
for light modulation. To the present 
time, unfortunately, no zinc sulfide crys- 
tals have been synthesized, and pure 
crystals of usable size are rarely found 
in nature. So, while the zinc sulfide 
crystal is the most desirable, it nonethe- 
less becomes necessary to study other 
crystals, particularly the types which can 
be easily synthesized in the laboratory. 
Two such crystals are ammonium dihy- 
drogen phosphate (ADP) and potassium 
dihydrogen phosphate (KDP). The 
ADP crystal has many commercial uses 



Fig. 2. Application of voltage 
to crystal structure. 

and is comparatively easy to grow. 
KDP crystals take longer to grow and 
do not have as many commercial appli- 
cations, although it will be shown that 
they do have certain advantages over 
the ADP for sound modulator applica- 
tions. 

Having stated an intention to use a 
crystal to produce a desired effect, it is 
necessary to state which section of the 
complete crystal will be used, since crys- 
tals may have different optical proper- 
ties for different sections through them. 
The section of the ADP crystal which 
displays the property we wish to study is 
a basal section, or a "Z" cut as it is 
termed. Figure 1 shows a complete 
crystal of ADP, and the dotted portion 
shows the basal section. Another con- 
sideration necessary to produce the de- 
sired effect is the application of the elec- 
tric field to the crystal section chosen, or 
how the crystal shall be submerged in the 
electric field. For zinc sulfide, the elec- 
tric field is applied perpendicular to the 
light beam passing through the crystals. 



206 



March 1953 Journal of the SMPTE Vol. 60 



With ADP, the electric field is placed 
parallel to the light beam, necessitating 
transparent electrodes. Nesa glass or a 
thin coating of chromium or gold makes 
excellent electrodes. Nesa glass (stannous 
chloride sprayed on glass), the most light- 
efficient, is the electrode used in the 
present modulators. 

An interesting property results when 
the mode of operation is such that the 
light beam and electric field are parallel ; 
namely, the light retardation depends 
only on the potential across the crystal. 
This is important since it makes the oper- 
ating voltage independent of the aperture 
of the optical system (Fig. 2). With a 
transverse electric field, the wider the 
aperture, the smaller the electric field in 
the crystal for a given applied voltage 
and therefore less retardation. If the 
electric field is parallel to the light pass- 
ing through the crystal,, the electric field 
is constant and independent of the aper- 
ture. This would seem to indicate that 
the voltage would depend on the thick- 
ness of the crystal. This is not the case, 
however, since the electric field is halved 
by doubling the crystal thickness, but the 
time, or length of path, the light beam is 
in the crystal is doubled and the two 
effects cancel. Therefore, the total re- 
tardation depends only on the potential 
placed across the crystal. Apertures of 
any size can be used and closed with a 
given potential, in contrast to the ex- 
tremely high voltage which would be 
necessary for a Kerr cell with large 
apertures. 

If the crystal thus described is placed 
between crossed polarizers and the zero 
potential condition examined, the result- 
ing transmission field, instead of being 
uniform, shows ellipses of alternate dark 
and bright sections. This configuration 
is the last phenomenon that needs in- 
vestigation in connection with ADP crys- 
tals. It is caused by the angular field of 
the beam of light entering the crystal. 
ADP not only retards the light beam 
when placed in an electric field, but it 
has a complex retardation function with 



incident light angle. That is to say, 
beams entering at different angles, in 
general, are retarded by different 
amounts. Consider the zero potential 
condition and refer to Fig. 3. The 
center black portion indicates retarda- 
tion for light that is normally incident to 
the crystal. Notice that for zero poten- 
tial applied, there is no birefringence; 
therefore, there is a dark spot in the 
center since only a single ray is passed 
and the polarizers are crossed. The 
next light ring on Fig. 3 indicates the 
locus of rays of all light entering at an 
angle away from normal incidence. 
The next dark section indicates a re- 
tardation of one ray with respect to the 
other of 360. So, while the light areas 
look alike and dark areas look alike, they 
nonetheless represent different amounts 
of relative retardation. It is for this 
reason that only the region near normal 
incidence is utilized in the modulator. 
This is accomplished by placing the crys- 
tal structure in a collimated beam. 
However, the conventional sound track 
has finite dimensions and it will be im- 
possible therefore to collimate the beam 
exactly. The effect of having to use a 
finite angular field will be discussed 
later. 

Transfer Characteristics 

With the device thus outlined, the first 
point of interest is that of the transfer 
characteristics, that is, a statement on 
voltage input versus light output. 
Figure 4 is a summary of the important 
relationships associated with analytically 
expressing the transfer characteristics. 
The function is not linear but it does have 
a reasonably linear portion. By expand- 
ing the transfer function about an arbi- 
trary bias point into a Bessel expansion, 
it is seen that harmonic distortions will 
occur due to the nonlinearity of the 
transducer as the percent modulation is 
increased. By choosing a proper bias 
point, the function can be made sym- 
metrical about the bias point, thus caus- 
ing all even harmonics to vanish. Figure 



Dressier and Chesnes: Electrooptic Recording 



207 




ANGLES OF INTEREST 



Fig. 3. Transmission field of crystal between cross polarizers with zero potential. 
208 March 1953 Journal of the SMPTE Vol. 60 



Intensity Transmitted: 



7- = COS 2 irt 
/O 



where t = retardation in waves. 
For an arbitrary bias point and sinusoidal modulation: 

I 1 , 1 r 
J~ ~ 2 2 S cos (0<s sm wt) sin 0o sin (05 sin o>/)] 

where 0$ = 2ir (retardation caused by modulated signal), 
and 0o = 2ir (retardation caused by the bias plate). 

Expanding in a Bessel Series: 

i \ ri -i 

j- = o ~f~ cos ^ 9 ^ Q ^ s ' ~f~ ^ 2 (^) cos 2ut + ^4(0^) cos 4<*)t + . . . 

sin 0o[^i(0s) sin ut + Js(6s) sin 3co + . . .] 

choose the bias plate 0o = ir/2 

.'. cos 0o = and all even harmonics vanish. 



Fig. 4. Transfer characteristic relationships. 



100 



75 



OPTIC 6/AS 



D. C. VOLTAGE APPLIED 
Fig. 5. Transfer characteristic. 



5 shows a plot of the transfer character- 
istics with a proper bias point. This 
point is labelled optical bias since it is not 
obtained by d-c biasing the crystal, but 
rather by inserting a mica quarter-wave 
plate between the polarizer and crystal. 
Substitution in these relationships indi- 
cates that for 74% modulation, the sec- 
ond harmonic distortion is zero, and the 



third harmonic distortion is approxi- 
mately 2.2% of the fundamental compo- 
nent. As the percentage modulation is 
increased, this figure rises rapidly, and 
thus the nominal modulation level for 
proper light valve operation is 75%. Of 
course, nonlinear electrical amplifiers 
may be included in the sound system 
allowing higher percentage modulation. 



Dressier and Chesnes: Electrooptic Recording 



209 



.75 
.70 
.65 
.60 
.55 
.50 
.45 
.40 
.35 
.30 
.25 
.20 
.15 
.10 
.05 

0. 











_\_ 


jjw 


)AM 


[NT 


1 




























































































































































































































2nd HARMONIC 
3[jj 3rd HARMO 


NIC 




DO' 


.01 .02 .03 .04* 
34'23" 1 C 8'45" 143'8" 217'31" 
ANGLE OF INCIDENCE 



RADIANS* 



.75 
.70 
.65 

.15 
.10 
.05 






























f FUNDAME 


NTAL 










































V 

^ 



, 


__ 
1 


,- 


! 







































































~JJ^ 


f 3rd HARMONIC 

"HARMONIC j __j H 


' 





.01 
34.'23" 



.02 

r8'45" 
ANGLE OF INCIDENCE 



.03 
T43'8" 



.04 
217'31" 



Fig. 6. Effects of angular field; crystal thickness, 0.045 in. 



7~"T 

ELI 

Fig. 7. Diagram of laminated structure. 



MICA 1 4 WAVE 
PLATE 



NESA - / 
ELECTRODES 



However, this technique has proven un- 
necessary, since the 75% level is sufficient 
for most needs. 

Angular Field 

The standard modulator slit has a 
relatively large dimension in one direc- 
tion, i.e., 0.1 in. Since this slit is the 
virtual light source for the modulator, all 
the light passing through the crystal 
cannot be normally incident. This 
angular field problem may be overcome 
by adjusting the length of the optical 
system and in that manner control the 
angular field. Figure 6 summarizes 
how the angular field affects distortion. 
For zero angular field and 74% modu- 



lation, the second harmonic distortion is 
zero; the third harmonic distortion is 
approximately 2.2% of the fundamental 
as noted earlier. As the angular field is 
increased, the third harmonic distortion 
remains substantially constant while the 
second harmonic distortion increases 
parabolically so that at 2.28 for the half 
angle, the total distortion from all sources 
is approximately 3.3%. To keep the 
modulator slit to the above-mentioned 
angular field, the overall modulator 
length is approximately 5 in. 

Frequency Response 

Before investigating the frequency re- 
sponse of a crystal modulator, it is well 
to study the laminated crystal structure 
as it is used. This is necessary since the 
laminar structure affects the frequency 
response. The term laminar structure is 
used since the polarizers, mica quarter- 
wave plate and crystal, with its elec- 
trodes, are all cemented together. This 
entire assembly is then mounted in a 
bakelite block. A schematic diagram of 
this laminar structure is shown in 
Fig. 7. Figure 8 shows a picture of this 
structure with a handling arm for fitting 



210 



March 1953 Journal of the SMPTE Vol. 60 







Fig. 8. Complete laminated structure. 



/ 5" 










fj 




M 






fKOP 




- 










fADP 




i K 

1 1 
/ 1 
^ -^' 1 


12Z 


Q 


EQUIVALENT i 


'/R.CUIT 






1 

1 

1 
1 

V.. 


&6 




C 4 C c 


C L 


















D 


': 




0.0 


^ Re 


RL 












/ 


3 AC 


>0 /C 


00 /d 


ooo too, 


000 



FREQUENCY IN CrCLCS 

Fig. 9. Frequency response of laminated ADP and KDP crystals. 



the laminar structure into the optical 
system. The frequency response for 
the ADP structure and its equivalent 
circuit diagram for the frequency region 
used for sound modulation are shown 
in Fig. 9. The structure does not 
respond to d-c since the crystal electrodes 
are cemented to the crystal, forming a 
capacitor resistance network (note 



equivalent circuit diagram). The d-c 
terminal resistance is approximately 
300 megohms while the terminal capacity 
is approximately 64 /x/zf (micromicro- 
farads). The series capacity made 
up by the electrode cemented to the 
crystal is quite large; therefore, the 
series impedance of these capacitors 
becomes negligible quite rapidly. As 



Dressier and Chesnes: Electrooptic Recording 



211 



Fig. 10. Schematic layout 
of light modulator optical 
system. 




' ZERO ADJUSTMENT RESISTOR A-C INPUT 
J 

MONITOR CHASSIS D-C SUPPLY CHASSIS 

Fig. 11. Interconnection diagram for complete audio system. 



indicated by the response curve, mid- 
frequency gain is attained from approxi- 
mately 3 cycles/sec until the first Piezo 
resonance, which in our instance is 
approximately 40 kc. This resonant 
point depends upon the physical size 
of the crystal, in particular upon the 
length of the sides of the crystal face. 
The frequency relationship is shown in 
the figure where K c for ADP is approxi- 
mately 110 kilocycle-centimeters. This 
constant varies slightly for the KDP 
crystal. Beyond the resonant peak, 
whose height incidentally depends upon 
the mechanical damping of the crystal, 
the sensitivity falls to about 0.6 of its 
original flat-region amplitude. The 
transition from one level to the next 
has been omitted from Fig. 9. It is 
not a smooth curve as shown in that 
figure. After transition, the response 
remains substantially flat until the next 
resonance which is a molecular resonance 
at extremely high frequencies. There 



have been crystals constructed where 
added mechanical damping has com- 
pletely depressed the resonant peak 
shown on the curve. Thus, a more 
uniform transition from one amplitude 
region to the other is obtained. These 
crystals, with proper electrodes, can be 
operated well into the megacycle regions. 
Figure 9 also shows a comparison of 
ADP and KDP. The frequency charac- 
teristics are very similar. Notice that 
the KDP is more sensitive (requires 30% 
less drive) to applied voltage than the 
ADP crystal. Also, the resonant peaks 
are not as pronounced for the same 
damping and further, the amplitudes 
after transition are very similar. 

Physical Description 

Figure 10 is a schematic representation 
for the layout of the optics of the sound 
modulator. A tungsten filament lamp 
was chosen as the light source since gas 
and arc lamps generate in their light 



212 



March 1953 Journal of the SMPTE Vol. 60 



beam a certain amount of noise which is 
undesirable. Between the tungsten coil 

| and the first optic, a blue filter is placed. 
This filter serves two purposes. The 
first is to limit the wavelength which 
will pass through the crystal structure. 
This is done since retardation is a func- 
tion of wavelength. Second, the filter 
acts as a heat-absorbing filter. For- 

| tunately, both of these requirements 
contribute only second-order effects to 
the output, and with no filter the increase 
in distortion is quite slight. However, 
since there is ample light available for 
our application, a blue filter has been 
included. The wavelength to be trans- 
mitted was determined by the spectral 
response of the motion picture film used. 
The lamp filament, which is of the 
tight-coil variety, is imaged slightly 
oversized and out of focus on a slit 
whose dimensions are 0.1 in. X 0.0006 
in. The filament image is run slightly 
out of focus to give a more uniform light 
distribution across the slit since the coils 
of the filament, even though they are 
wound quite closely, give some illumina- 
tion variation. The slit now becomes 
the light source and the next optic picks 
up this aperture and collimates the 
light beam for passage through the crystal 
structure. The laminar structure in- 
cluding the crystal has already been 
described. The placement of the slit 
pickup optic just mentioned is such 
that the angular field requirement is 
met. The final optic or projection optic 
gathers the collimated beam from the 
crystal and focuses the image of the slit 
onto the film. This optic is adjustable 
so that the image may be sharply focused 
on the film for different film distances. 
The magnification of the optical system 
is unity and therefore the modulation 
slit height remains 0.0006 in. high. 
This allows 9,000 to 10,000 cycles to be 
impressed on the motion picture film 
without amplitude loss when it travels 
at 90 fpm. The light transmission 
efficiency of the entire modulator is 16%. 



Electrical Requirements 

It was pointed out in an earlier section 
that the magnitude of the voltage for a 
given retardation is a function of wave- 
length. Since a filter has been inserted 
between the tungsten lamp and the 
crystal, the wavelength of the transmitted 
light is known. For 4500 A the required 
voltage is 7.4 kv per half-wave of re- 
tardation. The retardation required for 
75% modulation is something less than 
a quarter-wave and requires between 
1600 and 1800 v rms drive. It is 
interesting to note for zinc sulfide the 
driving potential is approximately 500 v. 

The electrical drive requirements for 
the ADP crystal structure can be deter- 
mined from the driving point impedance 
of the structure. The driving point 
impedance of these crystal structures is 
capacitive, and has a dissipation factor 
of approximately 0.07 which is sub- 
stantially constant to 50 kc. The ca- 
pacity measured at the crystal terminals 
is 64 /z/xf and is also substantially con- 
stant with frequency. The slight varia- 
tions from a constant as a function of 
frequency of these factors does not affect 
the design of the driving system and does 
not affect sound quality. The amplifier 
used to drive the modulator is a 50-w 
high-quality amplifier working into a 
special output transformer designed to 
drive this load over the frequency range 
desired. This design was accomplished, 
giving an overall system response essen- 
tially uniform from 30 cycles to 10,000 
cycles. 

Studies at this laboratory indicate 
that frequency components above 8500 
cycles are very often distorted during 
film processing and, in addition, many 
theaters have equipment which would 
only serve to distort frequency com- 
ponents above this figure. As a result, 
a wave filter is included in the electrical 
system to limit the nominal frequency 
passband from 75 cycles to 8500 cycles. 
The low end has been attenuated to 
filter out hum which may exist in some 



Dressier and Chesnes: Electrooptic Recording 



213 




Fig. 12. Completed modulator with cover removed. 




214 



Fig. 13. Modulator mounted on 35mm camera. 
March 1953 Journal of the SMPTE Vol. 60 



installations. If hum and the theater 
reproducing system do not cause low- 
frequency distortion, the low-frequency 
filter may be removed. A diagram of 
the electrical system is shown in Figure 
11. It indicates the required modula- 
tion equipment as well as the auxiliary 
equipment such as VI meters and 
calibrated attenuators. 

The lamp power supply is a variable 
d-c supply, feeding a 10-v 7|-amp lamp. 
With the 16% light efficiency figure, 
the lamp is run at 6 amp for positive 
recording and 6.8 amp for negative 
recording. Both levels are well below 
the nominal lamp rating. These values 
were determined from standard inter- 
modulation tests for various chemical 
baths. 

Summary 

The modulator described records 
sound on film at 75% modulation with 
a standard sound-modulation slit, and a 
total system distortion of approximately 
3.3%. The required driving voltage is 
2000 v rms derived from a standard 
audio amplifier through a suitable 
matching transformer. The frequency 
response of the entire system is sub- 
stantially flat from 30 cycles to 10,000 
cycles (i.e., within a quarter decibel). 
The entire modulator is extremely 
rugged and cannot be damaged by over- 
voltage. From an operations point of 
view, it's adjustment is exactly the same 
as the standard modulators now com- 
mercially employed for sound-on-film 
recording. 

A completed modulator with the cover 
removed is shown in Fig. 12. This 
figure shows the laminated structure 
being inserted into its proper position 
in the optical system. These crystals 
are interchangeable and can be changed 
in approximately five minutes. There 
are no figures available as to the life 
of these structures since none has failed 
to this date. One in New York City 
has been in constant operation since 
August 1950 and has been used over 



that period to record sound for approxi- 
mately 2700 hr. These crystals are 
stable solids and should have a life 
measured in tens of years. The crystal 
structure is held in place by spring clasps 
and can occupy only one position in the 
optical system. The tungsten lamp is 
mounted in a prefocus socket so that 
interchanging the lamp requires no 
adjustment of the optical system. Figure 
13 shows the modulator mounted on the 
35mm camera. The hole under the 
mounting is for a small fan which cools 
the entire housing to prevent crystal 
overheating and increase the lamp life. 

Acknowledgments 

The authors gratefully acknowledge 
the helpful suggestion of the Baird 
Associates Co., and in particular, the 
suggestions and criticisms of Robert 
Carpenter of that company. 

References 

1. Kerr Cell 

a. J. Kerr, "On a new relation between 
electricity and light," Phil. Mag., 4th 
Series, 50: 337-348, Nov. 1875. 

b. H. Abraham and J. Lemoine, "Dis- 
parition instantanee du phenomene 
de Kerr," Compt. rend., 729: 206-208, 
July 1899. 

c. F. Vilbig, Lehrbuch der hock frequenz 
Technik, Vol. II, 3rd ed., Edwards 
Bros., Ann Arbor, Mich., 1946, p. 
452. 

d. J. Kerr, "Electro-optic observations 
on various liquids," Phil. Mag., 5th 
Series, 8: 85-102, Aug. 1879. 

e. M. Born, "Elektronentheorie des 
natiirlichen optischen Drehungsver- 
mogens isotroper und anisotroper 
Fliissigkeiten," Ann. Physik, 55: 177- 
240, May-June 1918. 

f. E. F. Kingsbury, "The Kerr electro- 
static effect," Rev. Set. Instruments, 7: 
22-32, Jan. 1930. 

g. F. G. Dunnington, "The electro- 
optical shutter: its theory and 
technique," Phys. Rev., 38: 1506-1534, 
Oct. 1931. 

2. Pockels Effect 

a. F. Pockels, Abhandl. Ges. der I I'm. 
Gottingen, 39: \, 1893. 



Dressier and Chesnes: Electrooptic Recording 



215 



b. F. Pockels, Lehrbuch der Kristalloptik, 
B. G. Teubner, Leipzig, 1906. 

Bibliography 

Bruce H. Billings, "The electro-optic effect 
in uniaxial crystals of the type XH 2 PC>4. 
I. Theoretical," J. Opt. Soc. Am., 39: 
797-801, Oct., 1949. 
Ibid, "II. Experimental," 802-808. 

Robert O'B. Carpenter, "The electro- 
optic effect in uniaxial crystals of the 
dihydrogen phosphate type," J. Opt. Soc. 
Am., 40: 225-229, Apr. 1950. 

Discussion 

John G. Frayne (Westrex Corp., and Chair- 
man of the Session) : Is there any problem in 
the application of bias for noise reduction 
purposes? 

Mr. Dressier: This modulator will not 
work with conventional noise-reduction 
equipment but noise reduction is possible 
with special equipment designed for the 
crystal characteristics. No noise reduction 
is included with this system at the present 
time. 



Dr. Frayne: What is the signal-to-noise 
ratio on the film with this device? 

Mr. Dressier: I would guess in the 
neighborhood of 25 to 30 decibels. 

George Lewin (Signal Corps Photographic 
Center) : Can you take a picture of a subject 
through this device and use this principle 
as an instantaneous shutter. 

Mr. Dressier: I think scientifically the 
answer is yes. In fact, pictures have 
been taken using crystals of this type. 
The problem, however, is the narrow 
angular field. The crystals we are using 
now are 35 mils thick this gives a 5 
angular field of view. In photographic 
work, a wider field of view is generally 
required. This crystal is suited for use 
as a shutter with optical systems where 
you can control the light path so as to work 
with a nearly collimated light beam. 
When using a crystal of this type for 
photography, the object to be photographed 
is put a long way off so as to restrict the 
angular field. In general, the structure 
as we have it here is not suitable for 
picture taking. 



216 



March 1953 Journal of the SMPTE Vol. 60 



An Intermediate Positive-Internegatiye System 
for Color Motion Picture Photography 



By C. R. ANDERSON, N. H. GROET, C. A. HORTON and 
D. M. ZWICK 



A color printing system for motion pictures is outlined employing: Eastman 
Color Negative Safety Film, Type 5247; Eastman Separation Panchromatic 
Safety Film, Type 5216; Eastman Color Internegative Safety Film, Type 5243; 
and Eastman Color Print Safety Film, Type 5381. The sensitometric charac- 
teristics of the two intermediate materials are described. Some methods of 
using them and the problems involved in color registration printing are 
discussed. 



I 



N THE PRODUCTION of motion pictures 
in color, it is necessary, as in black-and- 
white work, to introduce special effects 
for dramatic emphasis or enhancement 
of the mood of the story. It is also 
convenient to be able to correct portions 
of the original negative footage for 
contrast, density or color balance because 
of unavoidable or accidental variations 
which may have occurred either in 
exposing or processing the original film. 
To accomplish these results it is con- 
venient to use duplicating steps which 
yield a color internegative whose color 
characteristics are such that it may be 
intercut with the original color negative. 
Such a sequence of steps is shown 
schematically in Fig. 1. The original 



Communication No. 1524 from the Kodak 
Research Laboratories, by C. R. Anderson, 
N. H. Groet, C. A. Horton and D. M. 
Zwick, Kodak Research Laboratories, 
Eastman Kodak Co., Rochester 4, N.Y. 
The paper was presented on October 18, 
1951, at the Society's Convention at 
Hollywood, Calif. 



camera negative may be printed directly 
to give a release print, or it may be 
printed through the separation positives 
and internegative to a release print. 
The first method is a rapid way of print- 
ing "dailies"; the second procedure 
gives an opportunity to introduce effects 
or adjustments in the final print. 

Such a duplicating system might be 
expected to consist of a color negative 
film, a color duplicating positive ma- 
terial, a color duplicating internegative, 
and a color release print film. This 
would follow the familiar practice in 
black-and-white work. Such a system 
has, at present, two practical disad- 
vantages. First, it would leave the 
producer's investment in production 
costs in dye images whose permanence 
or resistance to fading has not yet been 
proved. Second, the quality which can 
be obtained, at the present time, using 
four generations of color pictures, is 
not up to that which is obtained with 
the system illustrated in Fig. 1. 



March 1953 Journal of the SMPTE Vol. 60 



217 



Camera 


negative 






Separation 
positives 


Effects 






Corrections 




Color 






Inter negative 




PRINT PRINT 



Fig. 1. Sequence of steps involved in 
making color prints from a color negative. 



BLUE SEPARATION 
OR YELLOW PRINTER 
Exposed to tungsten through 
Kodak wrotten Filters N048A2B 
(intensity-scale sensitometer) 
Developed in Kodak SD-2iot70 F 




ZOO J.OO 

Relative log E 



1.2 



08 



GREEN SEPARATION 
OR MAGENTA PRINTER 

Exposed to tungsten through 
Kodak Wratten Filters No i6Noei 
(intensity-scale sensitometer) 
Developed in Kodak SD-21 at 




3.2 

28 
2.4 
20 



1.2 
08 



0.4 
0.0 



RED SEPARATION OR CYAN PRINTER 
Exposed to tungsten through 
Kodak Wratten Filter No 70 
(intensity-scale sensitometer) 
Developed in Kodak 
SD-21 Of 70F. 




2.00 

Relative 



.OO 

E 



2.00 1.00 

Relative log E 

Fig. 2. Families of characteristic curves of Eastman Separation 
Panchromatic Safety Film, Type 5216. 



The color system described in this 
paper is based on the use of four photo- 
graphic materials, three of these yielding 
dye images, one of them yielding silver 
images. These are: Eastman Color 
Negative, Type 5247; Eastman Separa- 
tion Panchromatic Film, Type 5216; 
Eastman Color Internegative Safety 
Film, Type 5243; and Eastman Color 
Print Safety Film, Type 5381. The 
first and last of these films have been 
described previously.* It is the purpose 
of this paper to discuss briefly the 
characteristics and properties of the two 
intermediate films just named and to 
outline the method of using them. 



* W. T. Hanson, Jr., "Color negative and 
color positive film for motion picture 
use," Jour. SMPTE, 58: 223-238, Mar. 
1952. 



Eastman Separation Panchromatic 
Film, Type 5216, is a black-and-white 
material, the sensitometric characteristics 
of which are shown in Fig. 2. The three 
sets of curves are time of development 
series for red, green and blue exposures. 
It will be noted that the contrast range 
is intermediate between that of the usual 
negative and positive materials in black- 
and-white. The emulsion is slower 
but gives sharper pictures than the 
usual panchromatic duplicating films. 
It may be processed in any standard 
black-and-white motion picture negative 
developer, the contrast being controlled 
by time of development, as shown in 
the curves. Figure 3 shows a wedge 
spectrogram printed below one on 
Eastman Fine Grain Duplicating Nega- 
tive, Type 5203. The sensitizing Jof 



218 



March 1953 Journal of the SMPTE Vol. 60 







Anderson, Greet, Horton and Zwick: Color Internegative 



219 



_>> 

I 

8 





3.8 



3.4 



3.0 



2.6 



2.2 



1.8 



1.4 



1.0 



6 




Relative log E 

Fig. 4. Characteristic curves of Eastman Color Internegative Safety Film, Type 5243; 
red, green and blue printing density versus relative log exposure. 



this new film extends to longer wave- 
lengths in the red region of the spectrum 
than that in the older material. This is 
to bring the maximum sensitivity close 
to the peak absorption of the cyan dye 
in the color negative, thus giving maxi- 
mum contrast and color separation for 
any specified time of development. The 
present film contains a dye in the 
emulsion which is not fully removed 
during processing. This dye imparts 
a slight greenish tint to the processed 
film. 

The internegative film, Eastman Color 
Internegative Safety Film, Type 5243, 
contains the same color formers and has 
the same density range as Eastman 



Color Negative Film, Type 5247. The 
speed, graininess and sensitization are 
quite different. Figure 4 shows the 
sensitometric characteristics of this film. 
The minimum densities of the blue and 
green are higher than that of the red, 
owing to the presence of the colored 
couplers. The contrast is higher than 
that of the negative film, Type 5247. 

In this film, the magenta coupler is in 
the top, blue-sensitive layer, so that the 
blue exposure leads to magenta dye. 
The green exposure results in the forma- 
tion of cyan dye, and the red exposure 
gives yellow dye. These noncomplemen- 
tary relations between sensitizer and dye 
image are used in order to profit from two 



220 



March 1953 Journal of the SMPTE Vol. 60 



Original 
Subject 

Eastman 

Color- 

Negative, 

Type 5247 



Original 




J Neg. Yellow Image 
Magenta Image 
Cyan Image 



Eastman 
Panchromatic 
Separatio 
Film " 
Type 5216 



Breen 

Separation Separation Separation 
or Yellow or Magenta or Cyan 

Printer 



Printer 



Printer 



Eastman Color 

Internegative Fi 

>24 



Type 5243 

Blue -Sensitive ^2 
Green-Sensitive 
Red-Sensitive 
Eastman 
Color Print 
Film, Type 5381 

Green- Sens itiv 
Red -Sensitive 
Blue-Sensitive 

Final Release 
Print 




Neg. Magenta Image 
Neg. Cyan Image 
**~~Neg. Yellow image 



Pos. Magenta image 
* Pos. Cyan Image 
*-Pos. Yellow Image 



r ig. 5. Schematic diagram showing relation of film sensitivity to dye formation through 
complete printing stages from color negative to the final release print. 



bserved facts: (1) that the magenta dye 
mage in a color film determines, to a 
arge degree, the sharpness of the picture ; 
nd (2) that the top layer of a multilayer 
film, in general, gives better picture sharp- 
ness than the lower layers. Of course, 
such relationships between color sensi- 
tivities and dye formation, if used in a 
camera film which was printed on a 
color film with normal sensitizing, would 
yield a false-color system quite useless 
for visual purposes. Because the three 
records of the separation positives are 
printed separately on the internegative, 
this condition is avoided. 

The internegative contains colored 
couplers, similar to those in Eastman 
Color Negative, which provide auto- 



matic masking to correct in part for the 
unwanted absorptions of the cyan and 
magenta dyes of the negative image. 
Thus, the duplicating stage does not 
introduce the brightness changes and 
hue shifts which would occur in the 
absence of this masking. 

Processing of the internegative is 
carried out in the same solutions and 
with the same precautions as Eastman 
Color Negative, though at reduced 
time of development. 

Figure 5 is a schematic diagram of 
the relations of sensitivity to dye forma- 
tion in the complete printing system 
from the color negative to the release 
print. At the top of the figure the red, 
green and blue represent the amounts of 



Anderson, Groet, Horton and Zwick: Color Internegative 



221 



these colors reflected from the original 
subject. These colors expose the camera 
film, forming the negative image in 
cyan, magenta and yellow dye, re- 
spectively. From the color negative 
three separation positives are prepared 
on Eastman Separation Panchromatic 
Safety Film. These positives are exposed 
through red, green and blue filters, 
giving a permanent silver record of each 
of the three dye images. It is now 
possible to print these records in the cor- 
rect dyes of the internegative film. This is 
shown in the lower portion of the dia- 
gram where the sensitivities appear on 
the left, the dyes formed on the right. 
Since it is necessary that the record of 
the dye images of the negative become 
finally the corresponding dye images 
of the internegative, it is obvious from 
the figure that the red-separation positive 
must be printed with green light. 
Similarly, the magenta dye in the blue- 
sensitive layer requires the green separa- 
tion positive to be printed with blue 
light, and the third, or blue, separation 
must be printed with red light. Thus, 
the original dye images of the negative 
will appear in the same dyes in the inter- 
negative and will give a print with 
normal color reproduction. 

In order to achieve proper tone re- 
production in printing the color inter- 
negative from the three separation 
positives, it is necessary to adjust the 
exposure so that the straight-line portion 
of the sensitometric curve of the color 
internegative is used. Underexposure 
at this stage causes loss of contrast in 
the shadows, as observed in the final 
print. Overexposure produces an inter- 
negative that may be too dense to print 
and may lead to low contrast in the 
highlight areas. 

The last stage in Fig. 5 shows the 
printing of the internegative on Eastman 
Color Print Film. This stage of printing 
may be done with white light modified 
as may be found necessary by Kodak 
Color Compensating Filters, or single 



or multiple light-source printers may 
be employed in which red, green and 
blue light are appropriately mixed at 
the printing aperture. 

We may turn now to a more specific 
study of these operations and to the 
precautions which must be observed. 
Because of the presence of the dye masks 
in the original negative, the choice of 
filters in making the separations is 
critical. The set of filters recommended 
for making these separations is: 



Separation 
Positive 

Red 

Green 

Blue 



Kodak Wratten 
Filter No. 

70 

16 + 61 
48A + 2B 



These filters have been chosen to give 
the highest contrast and the most satis- 
factory color separation from the masked- 
dye negative images. 

There are fewer restrictions on the 
choice of filters to be used when printing 
the separation positives on the inter- 
negative. Because the printing at this 
stage is from silver images, no effect 
is produced on the contrast or color 
reproduction by changing the trans- 
missions of the filters unless the trans- 
missions are made so broad that the 
filters transmit in an adjoining region 
of the spectrum. For example, the total 
transmission of the green filter must 
not be so broad that it exposes either 
the red- or blue-sensitive layer of the 
internegative film. The recommended 
filters have been chosen in this case to 
match the peak sensitivities of the inter- 
negative film as follows: 



Yellow printer or blue 

separation 
Magenta printer or green 

separation 
Cyan printer or red 

separation 



Kodak Wratten 
Filter No. 

29 

34 + 38A 
16 -f 61 



222 



March 1953 Journal of the SMPTE Vol. 60 



Shutters 




Gote 




- AXIS 



C 2 



Fig. 6. Schematic diagram of a two-lamp step-contact printer, of a type suitable for 
maintaining registration through two steps of printing without the necessity of chang- 
ing registration pins: L = lamps; C,, Cg = condenser lenses; A = heat-absorbing 
glass; F = filterholders. 



In order to produce a satisfactory 
final print with the color images in 
registration, two requirements in print- 
ing equipment must be met. 

1. Provision must be made for main- 
taining registration and correct orienta- 
tion during the printing of the separation 
positives from the original negative and 
during the printing of the internegative 
from the separation positives. 

2. Provision must be made for color 
timing. If an entire picture is to be 
printed from an internegative, it is likely 
that proper timing during the printing 
of the separation positives and inter- 
negative will result in an internegative 
from which the final release prints can 
be made at one printer light setting. 
However, if the final print is made from 
an intercut original and internegative, 
the printer used in making the final 
print will require color-timing equip- 
ment. Furthermore, the internegatives 
will require slightly different intensity 
timing since an internegative will be 
of somewhat greater density than the 
corresponding original negative. This 
may require an increase in printer 
intensity of several printer points. 

Because of the diversity of printing 
equipment it is impossible to give 
specific rules for every kind of printer, 
but some general suggestions may be 
useful. To maintain accurate regis- 
tration it is essential that the full pin 



of the registering printer enter the same 
perforation relative to the frame, both 
in making the separation positives and 
in printing these on the internegative. 
This may be achieved by using a step- 
contact printer with registration pins 
and two light sources, one on each side 
of the gate, as shown schematically in 
Fig. 6. In this figure, symmetrically 
placed on each side of the gate, are 
shown the lamps, L, the condenser 
lenses, C\ and C2, the heat-absorbing 
glasses, A, and the filterholders, F. In 
making the separations from the camera 
original, the lamp on one side is used. 
In printing the separations on the inter- 
negative, the lamp on the other side is 
used. In this way it is unnecessary to 
turn the separation positives over to 
maintain emulsion-to-emulsion contact 
in printing. Another way of accom- 
plishing the same result with a single 
light source is to have removable pins 
in the printing gate so that they may be 
changed to the correct position for the 
second operation. In the design of any 
registering printer it is desirable to 
have the full pin fill the same perforation 
as was used by the camera. This is not 
essential to registration when printing 
from an integral-tripack original but 
it improves screen steadiness. 

In such a step-contact printer, picture 
sharpness is improved if a partial 
vacuum is maintained between the two 



Anderson, Groet, Horton and Zwick: Color Internegative 



223 



c 


D : 




O! 








! 




M 




o 




D: 


A > 


o| 




S 




Q' 




OB 


B 


o 



Fig. 7. Diagram of a 
contact printing gate util- 
izing a partial vacuum 
for improved picture 
sharpness: A = printing 
aperture; B = registra- 
tion pins; C = channel 
to vacuum pump. 



3.8 



3.4 



3.0 



2.6 



2.2 



-S 1.8 

I 

I 1-4 



1.0 



.2 




Relative log E 

Fig. 8. Sensitometric curve of print from Eastman Color Negative Safety Film, 
Type 5247, as a neutral sensitometric scale on Eastman Separation 
Panchromatic Safety Film, Type 5216. 



224 



March 1953 Journal of the SMPTE Vol. 60 



films during the exposing part of the 
printing cycle. Figure 7 shows a front 
view of the contact printer gate, where 
A is the exposing aperture and B are 
the registration pins. Any air trapped 
between the two films is drawn out from 
holes matching the perforations by a 
vacuum pump attached at G through 
the channel, shown in dotted lines. 

If an optical printer is used for both 
steps, the same requirements must be 
satisfied regarding registration and may 
be achieved by the same means. If the 
separations are made on a step-contact 
printer and the internegative is printed 
on an optical printer, or vice versa, 
the location of the registration pins in 
both printers must be arranged to permit 
the maintenance of registration and 
orientation for projection. In any 
optical printing step, the resulting 
contrast will be dependent on the 
Callier Q-factor of the original. This is 
particularly true when printing from 
the separation positives because the 
silver image scatters far more light than 
a dye image does. This means, of 
course, that if separations have been 
made at the correct contrast for optical 
printing on the internegative they will 



give too low a contrast for a satisfactory 
contact print. 

In making certain types of effects 
using the present films, it must be kept 
in mind that the gammas of the three 
separation positives are not equal, since 
the unequal process gammas of the color 
internegative are compensated for in 
these positives. This is illustrated in 
Fig. 8, which shows the "print-through" 
sensitometric curves for a color negative 
neutral scale exposure printed on the 
separation positive film. The X's on 
the curves mark the densities of the 
whites and shadows occurring in a 
typical scene. The density ranges in 
the three separations are not equal. 
If a fade or dissolve is made by printing 
such positives on the internegative film, 
these density ranges must be spread over 
the same number of frames, or a shift 
in color balance will occur. Conse- 
quently, the shutter, diaphragm or 
other mechanism must be actuated at 
different rates for the different separa- 
tions. In cases where the effects can 
be introduced in making the positives, 
this difficulty is avoided because the 
original has matched gammas. 



Anderson, Groet, Horton and Zwick: Color Internegative 



225 



Kinescope Recording 
Film Exposure Control 

Bv RALPH E. LOVELL and ROBERT M. ERASER 



Various devices have been perfected to control accurately the exposure of 
video pictures to be recorded on motion picture film. These devices, com- 
bined with sensitometric control, take much of the guesswork out of the 
kinescope recording process. 



M, 



.UCH HAS BEEN WRITTEN about the 

various phases of recording television 
images on motion picture film. Equip- 
ment has been described, characteristic 
curves of the various components have 
been examined, and questions of resolu- 
tion have been discussed. Little ma- 
terial is available, however, about the 
methods by which proper film exposure 
can be obtained from a kinescope tube. 

It is the purpose of this paper to 
describe a practical operating procedure 
for obtaining optimum film exposure on 
kinescope recording negatives. This is 
a technique involving the use of some 
instruments well known in connection 
with electronic applications as well as 
others designed specifically for the 
purpose. 

One of the many handicaps under 



Presented on October 7, 1952, at the 
Society's Conventional Washington, D.C., 
by Ralph E. Lovell, National Broad- 
casting Co., Sunset and Vine, Hollywood 
28, Calif., who read the paper, and Robert 
M. Eraser, National Broadcasting Co., 30 
Rockefeller Plaza, New York 20, N.Y. 



which the technician worked in the early 
days of video recording was the lack of 
suitable instruments for determining 
conditions of film exposure. The early 
commercial kinephoto monitors had, 
as their sole instrument for determining 
the effects of changes in contrast, 
brightness and high voltage, a single 
50-jua meter, which by selective switching 
was supposed to indicate conditions in 
various circuits. The only method of 
judging maximum exposure conditions 
with this apparatus was to observe the 
average cathode current of the 5WP11 
recording kinescope. A rough approxi- 
mation of minimum exposure condi- 
tions referred to as black level 
could be obtained by an indication of 
the grid bias voltage of the 5WP11 
kinescope. With this crude method of 
instrumentation the technician was sup- 
posed to maintain uniform exposure and 
produce good photographic results. 
Needless to say, this was an almost 
impossible task. 

The situation today at NBC is quite 
different. Several instruments have 



226 



March 1953 Journal of the SMPTE Vol. 60 



100- 

80 - 

60- 

40- 

20- 



-20- 

-40- 




TYPICAL 

MEASURED 

LEVELS 



-*-WH1TE PEAK. 



REFERENCE 

WHITE LEVEL 



*- BLACK PEAK 

. - REFERENCE &LACK LEVEL* 
BLANKING LEVEL 



.-_ SYNCHRONIZING 
LEVEL 

NOTE: REFERENCE BLACK LEVEL TO BE SPECIFIED IN ACCORDANCE 
WITH OPERATING PRACTICES. 



Fig. 1. Television level measurements. 



been designed or adapted for this par- 
ticular purpose which have made ex- 
posure control a practical reality. 

Nomenclature 

Before describing these instruments 
a few of the terms used in television 
nomenclature may be briefly reviewed 
(Fig. 1).* Current television engineer- 
ing practice is to produce a composite 
video signal having a peak-to-peak 
amplitude of 1.4 v. The reference zero 
for a composite signal is at blanking 
level, as shown at the left of Fig. 1. 
The synchronizing pulse is required to 
have an amplitude of 0.4 v or 40 units 
in a negative direction, from the reference 
zero. Picture information is contained 
in the 100-unit portion above reference 
zero, having a maximum of 1 .0 v. Thus 
peak whites represent 1.0 v above zero 
with black picture information com- 
monly referred to as black level falling 
about 10 units or 0.1 v above zero. 

Exposure Control Equipment 

Since 1.4-v peak-to-peak amplitude 
has been adopted as standard for video 
signals it is essential that some reliable 
source be available with which to cali- 



* J. H. Roe, "Standardizing and measuring 
video levels in a TV station," RCA Broad- 
cast News, 65: 30, July-Aug. 1951. 



brate monitors and oscilloscopes through- 
out the television plant. One such 
source is a peak-to-peak reading volt- 
meter which is connected across the 
output of a step-down 60-cycle trans- 
former and potentiometer. By care- 
fully adjusting this meter to read 1.4 v 
and by applying this voltage to the 
"calibrate" position on all monitors in 
the television plant it becomes possible 
to obtain identical levels from the 
various studios. A uniform calibrating 
voltage, such as this, is of extreme im- 
portance in kinescope recording and 
forms the basis for all accurate measure- 
ment. 

The most important single instrument 
for controlling exposure is the step 
generator (Fig. 2), a tool which possesses 
great value and versatility. This is an 
electronic signal generator designed and 
built by NBC principally for kinescope 
recording, although it is useful in other 
parts of the television plant and par- 
ticularly, in this instance, for the stepping 
of the video signal level. When hori- 
zontal-drive, synchronizing and blanking 
signals are fed from an external source, 
the generator is capable of producing 
four types of signals (Figs. 3, 4, 5 and 6). 

The step signal, as seen on the face 
of a kinescope (Fig. 3), resembles a 
photographic step tablet such as that 



Lovell and Fraser: Kinescope Exposure Control 



227 




Fig. 2. Step generator in rack adjacent to TM-5A master monitor. 



used in intensity-scale sensitometers. 
The number of steps can be adjusted, by 
means of a potentiometer, from 3 to 22. 
However, it is usually set to produce 10 
steps, simply because the resulting 10 
density strips on each frame of motion 
picture film are large enough to be read 
easily by a densitometer. Step 1, read- 
ing from the bottom, is adjusted to simu- 
late black level, i.e., 10 units above 
blanking level. Each succeeding step 
raises the signal 1 units above its prede- 
cessor until step 10 represents 100 
units or peak white. 

A second type of signal (Fig. 4) 
produced by the step generator is a 
rectangular pulse, approximately 10 
fjisec in width, and occurring at the 
horizontal scanning rate of 15,750 
cycles/sec with an amplitude controllable 
from blanking level to peak white. A 
pedestal control can be adjusted simul- 



taneously to simulate black level while 
the pulse is adjusted to 100 units. In 
this way the extremes of picture in- 
formation relating to a given video 
signal may be represented. If desired, 
the pulse may be eliminated and the 
pedestal control adjusted to any level 
between zero and 100 units, thereby 
producing a blank raster of uniform 
brightness dependent upon the pedestal 
setting. 

An external input position permits the 
introduction of various types of test 
signals, such as the 4,500-cycle sine 
wave shown in Fig. 5, or the 15,750-cycle 
sine wave seen in Fig. 6. 

An oscilloscope with a 7-in. calibrated 
screen plays an important part in ex- 
posure control, for not only can it be 
used in conjunction with the step 
generator signals to determine proper 
exposure settings for the 5WP11 kine- 



228 



March 1953 Journal of the SMPTE Vol. 60 



scope, but it can also monitor the 
television picture signals being re- 
corded and immediately indicate to the 
operator discrepancies in level which 
might need correction. The 7-in. screen 
is preferred because it presents an 
expanded scale upon which level varia- 
tions can be readily detected. 

An ingenious little device used to 
advantage by our staff is a spot photom- 
eter made in England by Salford 
Electrical Instruments, Ltd. Although 
not currently in operational use in 
NBC Hollywood it has served well to 
establish the control procedure under 
discussion and might well serve other 
interested persons. Brightness values 
in log foot-Lamberts can be read directly 
from small areas on the face of the re- 
cording kinescope. Since visual match- 
ing of the luminances of fields of different 
colors presents difficulty, it is desirable 
to measure the photo-actinic output of 
the phosphor; a blue filter may also be 
installed in the photometer as an aid in 
matching the kinescope phosphor color. 
A calibrated density wedge, incorporated 
in the instrument, is visually compared 
through an eyepiece with the brightness 
produced by the small area being 
measured. Operation, in this respect, 
is much like that of a Capstaff-Purdy 
Densitometer. Since this photometer 
can read minute areas on the face of the 
kinescope it can be used to determine 
the brightness of the various steps of the 
step generator signal, particularly the 
two extremes, step 1 and step 10. Kine- 
scope brightness may therefore be set in 
accordance with predetermined bright- 
ness values to produce desired exposure 
on the film negative. 

While this is a very useful instrument, 
it has three operational disadvantages: 

1. The human element enters into 
the readings, hence no two people will 
get precisely the same results. 

2. Due to slight differences in kine- 
scope phosphor color, it is often difficult 
to decide when the two comparative 
densities are perfectly matched. 



3. Calibration drift occurs rapidly 
due to insufficient current capacity in 
the exposure lamp supply. One remedy 
for this is an auxiliary battery box 
supply. 

Use of the spot photometer soon led 
to the design and construction of a 
different device eliminating the human 
element from the kinescope brightness 
measurements. This was the phototube 
amplifier shown attached to the hood 
in Fig. 7 which, although not impressive 
in appearance or complexity, has intro- 
duced a high degree of accuracy into 
our exposure control measurements. 
Positioned just outside the recording 
light path, the type 929 high-vacuum 
phototube receives light from the kine- 
scope face during scanning and no light 
during the horizontal and vertical retrace 
times. 

If employed only with a blank raster 
produced by the step generator this 
phototube and its amplifier produce an 
electrical waveform having an amplitude 
proportional to the intensity of the light 
output from the kinescope. By frequent 
calibration of the phototube amplifier 
an exact comparative measurement of 
light output can be obtained and fed to 
the calibrated 7-in. oscilloscope (Fig. 8). 
A low-gain switch position is used for 
peak-white measurements where much 
light is produced by the kinescope, and 
a high-gain switch position is provided 
for black-level measurements where a 
very low level of light output is produced 
by the kinescope. Thus the two ex- 
tremes of exposure, namely peak white 
and black level, can be accurately set 
with no "human element" error such 
as occurred with the spot photometer. 

In addition to the electronic instru- 
ments thus far mentioned, two con- 
ventional instruments well known to 
the film industry are employed. These 
are the Eastman Processing Control 
Sensitometer, an intensity-scale type, 
and the Eastman Densitometer, known 
generally as the Capstaff-Purdy Densi- 
tometer after its inventors. 



Lovell and Fraser: Kinescope Exposure Control 



229 





230 



March 1953 Journal of the SMPTE Vol. 60 




H 

i i 

. v 



t! 

V - 



<0 

bo 




br, 

a 



Lovell and Fraser: Kinescope Exposure Control 



231 



Operating Procedure 

The operating procedure used to 
maintain accurate exposure control with 
the instruments described above may 
be classified in four categories: 

1 . Calibration of equipment. 

2. Peak-white measurements. 

3. Black-level measurements. 

4. Sensitometry. 

Calibration of Equipment. As indicated 
earlier, it is extremely important that 
all units in the recording plant be 
carefully calibrated, since amplitude 
drift or level discrepancies affect film 
exposure. The five operational steps 
for calibration are as follows: 

(1) Adjust the 60-cycle calibrating 
voltage to exactly 1.4 v peak-to-peak 
as read on the meter previously de- 
scribed. 

(2) With the oscilloscope selector 
switch hi the "calibrate" position, 
adjust the vertical amplitude of the 
TM5A master monitor oscilloscope until 
the 1.4-v calibrating signal exactly fits 
arbitrarily prescribed lines for the 1.4-v 
video signal. 

(3) Apply the 1.4-v calibrating signal 
directly to the input of the 7-in. oscillo- 
scopes at each recording position and 
adjust amplitude to prescribed lines on 
the tube face. 

(4) Apply the 1.4-v calibrating signal 
to the input of the phototube amplifier 
and adjust its gain until the output as 
seen on the scope fills prescribed cali- 
brating lines. 

(5) All electronic measurement instru- 
ments in the system have now been 
calibrated. 

Peak-White Measurements. There are 
several means by which peak-white 
measurements can be made satisfac- 
torily. From an operational standpoint 
the most desirable is as follows: 

(1) The familiar step signal produced 
by the step generator is used and the 
three important steps, 1, 2 and 10, 
carefully adjusted to prescribed lines 
on the face of the master monitor 
oscilloscope. 



(2) This signal is then fed to each re- 
cording position through exactly the 
same distribution amplifiers and co- 
axial lines as will be used for the picture 
signal to be recorded later. The 7-in. 
oscilloscope is then connected across 
the output of the recording video ampli- 
fier, and the amplifier gain adjusted to 
the desired amplitude as shown on the 
calibrated oscilloscope. 

(3) There is, of course, a fixed rela- 
tionship between signal amplitude and 
kinescope light output, hence also 
maximum film exposure. 

Black-Level Measurement. A somewhat 
different procedure is employed for 
black-level measurement since it has 
been found to be much more critical 
than peak-white measurement. 

(1) The blank raster, or pedestal, 
position of the step generator is used, 
carefully adjusted to 20 units on the 
master monitor oscilloscope. The 20- 
unit adjustment, instead of the customary 
10 for black level, was chosen because 
it produced a density of about 0.1 on 
the film, a value far enough removed 
from the base fog region to give more con- 
sistent operating data. 

(2) The 7-in. oscilloscope at each re- 
cording position is now connected to 
the output of the phototube amplifier. 
The black-level control is then adjusted 
to produce the desired amplitude as 
indicated on the calibrated oscilloscope. 
This, of course, will determine the 
minimum exposure, or black level, on 
the film. 

The two extremes of the anticipated 
video signal, peak white and black level, 
have now been set. 

Sensitometry. The final test of the 
above is to make an exposure, develop 
the film, and measure the resulting 
densities. 

(1) This is done using the step signal 
carefully adjusted as described. 

(2) The test exposure is made about 
two hours before recording time to 
permit development, analysis and re- 
adjustment. 



232 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 7. Phototube amplifier mounted in operating position on kinephoto monitor. 



929 
PHOTOTUBE 




1.4 VOLT 

PEAK-TO-PEAK . 
CALIBRATING / 
VOLTAGE 

I 
I 

L 



Fig. 8. Schematic diagram of phototube amplifier. 
Lovell and Fraser: Kinescope Exposure Control 



233 



(3) An intensity-scale sensitometric 
exposure is made on each film recorded, 
whether test or show, to determine 
processing conditions at time of develop- 
ment. 

(4) Step generator densities, par- 
ticularly those resulting from steps 1, 
2 and 10, are read and brought to the 
attention of the operator at each camera 
position. If slight adjustments in ex- 
posure are indicated the operator makes 
them just prior to recording time. 

(5) A short section of step signal is 
recorded either at the head or at the tail 
of each show. An intensity-scale sensi- 
tometric exposure is also made. 

(6) A characteristic curve is plotted 
for each show as well as a step density 
plot resulting from the step generator 
exposure. 

(7) The data thus accumulated are 
recorded in an operating log maintained 
at each camera position as well as in our 
sensitometric file. 

The above procedure, described in 
its necessary detail, may sound laborious 
and time consuming. Actually the whole 
process can be accomplished in a few 
minutes, with spot checks requiring less 
than 30 sec. Consistency of exposure, 
over long periods, resulting from this 
method is ample testimony to its value, 
and we are convinced that the quality 
of our product is greatly enhanced by 
its use. 



Discussion 

John G. Stott (Du-Art Film Laboratories}: 
Can you tell us how you take care of the 
problem of resolution of scanning lines 
when you do the densitometry of that 
step-wedge tablet? 

Mr. Lovell: When we read the step- 
wedge with the densitometer? Well, act- 
ually we ignore it. We realize that 
probably there are some strange things 
going on there, but we haven't tried to 
analyze what the effects really are. We 
do get a perfectly satisfactory density 
reading and while from a theoretical 
standpoint it may not be what it would be 
on a normal film, nevertheless it is con- 
sistent and it does give us operating data 
which are very useful. 

Mr. Stott: Well, you have to do your 
densitometry with a visual instrument 
then, is that correct? 

Mr. Lovell: We use the Eastman Cap- 
staff-Purdy Densitometer to make all 
our readings. 

Mr. Stott: You would probably run into 
some difficulties then if you used the elec- 
tronic or photocell type of densitometer? 

Mr. Lovell: I believe not. We have used 
a regular ERPI Densitometer with no 
detrimental effects. We get comparable 
readings and, of course, they don't read 
exactly the same, but no two densitometers 
ever do, I believe, and as long as there is 
a constant difference between the two, we 
are happy. I should say that perhaps 
two ERPI Densitometers do read the 
same, but not necessarily two different 
types of densitometers would read the 
same. 



234 



March 1953 Journal of the SMPTE Vol. 60 



Time-Zone Delay of Television 
Programs by Kinescope Recording 



By RALPH E. LOVELL 



A technique is described whereby network television programs can be delayed 
for three hours to compensate for the time differential between the east and 
west coasts. The use of 35mm film for picture and magnetic recording for 
sound insures high-quality reproduction. 



I 



N TELEVISION, as in radio, the matter 
of a three-hour time difference be- 
tween the east and west coasts of the 
United States creates a serious problem 
in releasing programs at a time which will 
reach the desired audience at all points 
throughout the country. In radio, the 
problem has for years been solved by disc 
or tape recording with appropriate hours 
of delay in playback to compensate for 
the time-zone differential. The opening 
of the coast-to-coast television network 
in 1951 created a similar challenge to 
kinescope recording of television pro- 
grams. 

As in radio, there are two aspects of 
this delayed broadcasting. Many live 
television programs intended primarily 
for eastern evening reception originate in 
Hollywood in the late afternoon, but are 
not released to the west coast cities at the 
time of origination because the desired 
western audience is not available at that 

Presented on October 7, 1952, at the 
Society's Convention at Washington, D.C., 
by Ralph E. Lovell, National Broadcasting 
Co., Sunset and Vine, Hollywood 28, Calif. 



time. These programs can now be re- 
corded in Hollywood and reproduced 
usually three hours later for viewers in 
Los Angeles, San Francisco, Seattle and 
San Diego, at the west coast evening 
hour desired by the sponsor or network. 
Conversely, many shows originate in 
New York, or other eastern cities, at a 
time convenient to eastern audiences, 
but again they are three hours too early 
for western audiences. These programs 
can likewise be recorded in Hollywood at 
the time of their live origination in the 
east and be telecast throughout the west 
coast three hours later. The terms 
"Quick Kine" or "Hot Kine" have been 
coined to describe such three-hour delay 
recordings, the techniques for accomplish- 
ing which are new and rather complex, 
and depend upon close coordination be- 
tween recording, processing, editing, 
projection and messenger personnel. 

Figure 1 shows a portion of the kine- 
scope recording room at NBC Holly- 
wood. On the right is a 16mm single- 
system camera, and on the left a 16mm 
sound recorder of conventional design. 



March 1953 Journal of the SMPTE Vol. 60 



235 




Fig. 1. A portion of the kinescope recording room at NBC Hollywood. 

On the right is a 16mm single-system camera, on the left a 16mm sound 

recorder, and in the center a 35mm camera. 




236 



Fig. 2. A 35/16 synchronizer with a magnetic reproduction 
head for editing purposes. 

March 1953 Journal of the SMPTE VoL 60 



In the center is one of the three 35mm 
cameras installed about a year ago to 
accomplish the three-hour delay record- 
ing to be described here. It is to be 
noted that the 35mm camera is equipped 
with a 3000-ft magazine, permitting a 
running time of 33 min. The 35mm 
cameras used thus far have been of the 
"silent" type; that is, they record only 
the picture portion of the program. It 
is expected that single-system 35mm 
cameras capable of recording both pic- 
ture and sound will soon replace the 
silent cameras shown here. The sound 
portion of the program is currently being 
recorded very successfully on 1 6mm per- 
forated magnetic film, running at the 
1 6mm speed of 36 fpm. For this purpose, 
two rack-mounted, synchronously driven 
RCA magnetic recorder-reproducers are 
employed. Two RCA 16mm optical 
recorders which have been modified by 
the addition of a recording head and a 
monitor head may also be used to per- 
form magnetic recording. Lip syn- 
chronization between picture and track 
is cued by insertion of an audio tone in 
both picture monitor and sound recorder, 
thereby creating easily recognizable pat- 
terns comparable to studio slating marks. 
It has been the experience of broad- 
casters for many years that a small 
investment in an extra copy of a disc or 
tape is amply repaid because some acci- 
dent may damage the "A" copy record- 
ing. This same philosophy applies to 
kinescope recording: hence most com- 
panies make two negatives, one being 
considered the "A" copy, the other the 
protection, or "B" copy. It is in this 
quick kine "B" copy operation that 
NBC's single-system, 16mm cameras are 
particularly useful, for they permit the 
recording of both a picture negative and 
a photographic, variable-area, direct- 
positive, sound track on one, single-per- 
forated strip of film. At the end of each 
half-hour of quick kine recording, both 
the 35mm and 16mm magazines are 
immediately removed from the cameras 
and taken into the darkroom for unload- 



ing. It is to be noted that only two and 
a half hours now remain till the film must 
be processed, edited, and in the projector, 
ready to be rebroadcast to the west coast 
audience extending from San Diego to 
Seattle. 

The 16mm single-system "B" copy is 
developed by NBC personnel in one of 
two Houston Model 22 processing ma- 
chines. A half-hour program requires 
about 45 min for complete processing, 
including a waxing treatment to enhance 
projection. Meanwhile, the 35mm "A". 
copy negative has been placed in its 
original light-tight carton, and handed 
to a waiting messenger who drives a few 
blocks across town to a commercial film 
laboratory where a crew is prepared to 
process the film immediately. Since the 
film used for the kine camera negative is 
actually a low-contrast positive, it runs 
through positive developer at the normal 
machine speed of 125 fpm, resulting in a 
negative gamma of about 1.60. It 
therefore creates no special problems, 
such as changes in machine speed or 
threading, and can be connected directly 
to other positive work going through the 
machine. After the film passes through 
developer, hypo and wash, it receives a 
waxing treatment which lubricates the 
film to permit immediate projection. 
Approximately one hour after the 2700-ft 
roll of undeveloped film reaches the 
laboratory, it comes off the dry end where 
the messenger is waiting to return it to 
the television station for editing. 

During the time the two negatives are 
being developed, the editing personnel 
are not idle. With the aid of a magnetic 
sound reader, the synchronization tone- 
bursts at the head and tail of the 1 6mm 
magnetic film are located and marked 
with a wax pencil . Identification leaders 
are prepared in advance and attached 
to the All-Purpose Film Leader which has 
replaced the Academy leader for tele- 
vision films. The 16mm single-system 
"B" copy negative then arrives from 
processing, and is made ready for projec- 
tion by the addition of the prepared 



Ralph E. Lovell: Kinescope Time- Zone Delay 



237 




Fig- 3. Quick kine projection facilities; a 35mm projector with an interlock 
motor which drives the 16mm magnetic film reproducer in the background. 



leaders. The return of the 35mm "A" 
copy negative is the signal for prompt 
action by the editing personnel who 
place the film on a 3000-ft, double- 
sided flange and proceed to line it up in 
the synchronizer with the 16mm mag- 
netic sound track. At the request of 
NBC, the Moviola Company installed a 
magnetic reproducing head in one of the 



16mm wheels of a 35/16 synchronizer 
(Fig. 2), thereby making it a combina- 
tion sound reader and synchronizer, and 
permitting rapid and safe handling of 
the two films. Identification leaders 
are attached, a special adhesive marker is 
applied to the magnetic track for accurate 
thread-up positioning, lip synchronism is 
checked at both ends of the show, and 



238 



March 1953 Journal of the SMPTE Vol. 60 



the 35mm negative is mounted on a 
3000-ft projection reel. All of the above 
processing and editing, including trans- 
portation of the 35mm negative across 
town twice, has taken approximately 
two hours, leaving about a half-hour for 
threading in the film projection studio. 

Close examination of Fig. 3 reveals a 
selsyn interlock distributor motor, chain- 
driven by the RCA 35mm television pro- 
jector motor. This interlock motor 
supplies driving power to one of the RCA 
rack-mounted 16mm magnetic repro- 
ducers located at the right of this picture. 
The 1 6mm magnetic track therefore runs 
in perfect lip synchronism with the 35mm 
picture negative. By simply flipping the 
iconoscope polarity switch to "Negative" 
and reshading accordingly, a positive 
picture image of high quality is obtained 



from the negative film. The single-sys- 
tem "B" copy is run in a 16mm projec- 
tor on an adjacent iconoscope, so that in 
the event of trouble with the "A" copy, a 
flick of a button will put the "B" copy 
picture and sound on the air. 

If 1 6mm release prints are required for 
syndication, they are obtained by reduc- 
tion printing from the "A" copy 35mm 
picture negative and by contact printing 
from a 16mm negative sound track. 
Picture quality thus obtained on 16mm 
release prints is superior to that obtained 
by contact printing from a 16mm nega- 
tive of the same subject matter. 

Although this three-hour time zone 
delay operation may sound very pre- 
carious and nerve wracking, it has been 
in daily use for one year with very few 
errors and with picture and sound 
quality rivaling that of many live shows. 



Ralph E. Lovell: Kinescope Time- Zone Delay 



239 



History and Present Position of 
High-Speed Photography in Great Britain 



By W. DERYGK GHESTERMAN 



The history of high-speed photography in Great Britain is outlined, beginning 
with intermediate-rate cameras, defined as those in which the film is trans- 
ported continuously through the camera mechanism at speeds not exceeding 
40 m/sec. Then follows a survey of drum cameras, in which a single loop of 
film is transported at speeds up to 240 m/sec on a rotating drum, or where 
images are swept along the stationary film at this rate by moving optical 
parts. Recent developments in light sources of short duration are discussed 
and the review concludes with a description of some research studies in 
zoological, biological and medical sciences, and some recent military applica- 
tions. 



J_ HE TERM HIGH-SPEED PHOTOGRAPHY 

is now generally used to cover a wide 
range of observational techniques in 
which photographic images are recorded 
to examine the sequence of happenings 
in a rapid event. Sometimes a series of 
pictures is produced for subsequent 
cinematographic projection, or alterna- 
tively, separate individual pictures per- 
mit study of a repetitive motion of a 
repeatable event, by subsequent anal- 
ysis. It is convenient to classify the 
methods both by the repetition rate of 
taking pictures and by the individual 
picture exposure time. Each of these 
fundamental variables is of vital im- 
portance to the camera designer and to 

Presented on October 9, 1952, at the 
Society's Convention at Washington, D.C., 
by W. D. Chesterman, Royal Naval 
Scientific Service, Admiralty, London, 
England. 



the research worker who uses the 
observational tool which has been 
devised. The progress of high-speed 
photography in Great Britain, as in 
Europe and the United States, may be 
considered as a search for methods of 
higher repetition rate together with 
ever shorter exposure times. It is always 
most desirable, but also difficult, to 
maintain the highest picture quality 
while improving these other variables. 

Intermediate-Rate Cameras 

In the present paper we shall deal 
first with intermediate-rate cameras, or 
those in which the film is transported 
continuously through the camera mech- 
anism at speeds of less than 40 m/sec. 
These cameras are of two categories 
those in which the film is moved inter- 
mittently and is stationary at the time 
of exposure, and those in which the film 
moves continuously at a high rate, the 
images being optically stabilized on the 



240 



March 1953 Journal of the SMPTE Vol. 60 



(film by a variety of means. The author 
has reviewed elsewhere 1 the variety of 
instruments made in Great Britain, 
Germany, France and the U.S.A. 
during the last 30 years. Unlike some 
cameras in the higher ranges of speed, 
many of these instruments were pro- 
duced commercially in numbers of a 
I few dozen, and in rare cases, of a few 
hundred. In Great Britain the em- 
phasis has been on types using 35mm 
rather than 16mm film. This may well 
be due to the feeling among many 
workers that the large picture size is a 
great aid to certain detail analysis, and 
for this reason the higher repetition rate 
possible with smaller format films (16mm 
and 8mm) has been sacrificed. 

One may instance three instruments. 
For speeds up to 240 or 300 frames/sec, 
the Vinten HS. 300 is a valuable and 
flexible camera. 2 It is in the category 
of a normal slow-motion cine camera, 
and has the facilities normally associated 
with fine-quality studio cameras. It 
has been widely used in field work and 
has proved capable of producing a 
picture quality equal to that of the best 
cine cameras. The maximum repetition 
rate at which it can be used is 10 times 
normal cine-frame rates. Beyond this, 
cameras using a different film-transport 
system are used, the film motion being 
continuous rather than intermittent. 
Optical stabilization may be achieved 
by a lens disk of 48 matched objectives, 
and Geary 3 has described the Vinten 
HS. 2000 camera which was produced 
during the last war for ballistics research. 

A more recent development in 35mm 
cameras was constructed by Myers and 
the author 4 in 1950, at the firm of 
Acmade Ltd. 

In this instrument, normal picture 
formation is stabilized on the film by a 
spinning cubical prism of the con- 
ventional form. In order to achieve 
a powerful film drive coupled with high 
acceleration to full speed, the design 
is based on a large-diameter 24-frame 
sprocket, and the film is in contact over 



half the circumference. The sprocket 
is driven from the prism shaft with 
appropriate reduction gearing. The 
camera takes 2000 normal-size 35mm 
frames/sec, ready for cinematographic 
projection. With an octagonal-section 
prism the maximum rate becomes 4000 
frames of half height per second. With 
the prism removed the camera can be 
used as an oscillograph recorder with 
a film-transport speed of 40 m/sec. 
The optical systems embodied in the 
camera permit the simultaneous trigger- 
ing of a power stroboscope, synchronized 
with the picture formation. In a direc- 
tion at right angles to the main picture- 
forming axis, a collimated light beam 
passes through the spinning prism on tq a 
photocell to trigger the light source of a 
short-duration flashtube. It is also 
possible to mask off a part of the picture 
area and then image the spot deflection 
of a high-brightness oscilloscope onto a 
part of the rapidly moving film. The 
normal timing markers at 1000 cycles 
are recorded on the film rebate. 

This camera has been widely used 
during 1951 and 1952, in Great Britain, 
for cavitation studies, and for various 
ballistic research. It is a convenient 
instrument for photomicrographic studies 
at high speed. The stimulus and 
achievement of similar American de- 
velopments is readily acknowledged. 

In Great Britain there has not yet 
been constructed, to the writer's knowl- 
edge, a 16mm camera in the inter- 
mediate speed range. This is probably 
due to the availability of American, 
German and French instruments, which 
have often been in small quantity 
production. 

Drum Cameras 

At rather higher picture repetition 
rates, it is no longer possible to transport 
film at the necessary speeds continuously 
through a camera mechanism. Drum 
cameras have been devised to take 
pictures at rates from 2000 to 200,000 
pictures/sec, depending on the picture 



W, D, Chester-man: High-Speed Photography in Great Britain 



241 



size and the elaboration of the instru- 
ment. The earliest British type was 
termed the Heape and Grylls Machine* 
and was described by Connell 5 in 1926. 
At that time, a quarter of a century 
ago, it was a considerable optical 
achievement. By rotating a 1.73-m 
drum at 1000 rpm a transport rate for 
the film of nearly 100 m/sec was ob- 
tained. Optical image formation and 
stabilization were carried out by a disk 
of lenses rotating in front of the film 
plane. The camera could take stereo- 
scopic pairs of pictures and a picture 
repetition rate of 5000/sec was obtained. 

In those days of very slow emulsions, 
lighting the subject presented a pro- 
digious problem, and two searchlights 
were used to focus the light energy 
(and heat!) on the subject. Flash dis- 
charge sources or pulsed arcs make the 
present-day problem far easier to solve. 
The Jenkins camera 6 was of the same 
period and used the same principle of 
optical compensation. 

Altogether simpler instruments were 
made for specialized tasks by Baxter 7 
and Brailsford. 8 The latter used a 
rotating cubical prism block and a 
simple loop of film on a small drum. 
Fifty successive pictures of the event 
were recorded, and the cameras were 
eminently suitable for self-luminous 
subjects such as the changes in a welding 
arc. 

The Second World War was the stimu- 
lus to more elaborate constructions. 
Notable amongst these was the Sco- 
phony Camera, 9 which, unlike the earlier 
four cameras mentioned, was produced 
commercially in small numbers. This 
instrument was one of the first to succeed 
in combining the high repetition rate of 
10,000 pictures/sec with a large image 
size of 24 X 24 mm. A mirror drum 
with 49 facets rotated concentrically 
with the film drum to achieve image 
stability during exposure. A film drum 
of 0.39-m diameter rotated at 12,000 
rpm to give a film-transport speed of 
240 m/sec. 



A camera of quite different design, 
the Marley camera, has already been 
reviewed in this Journal by Shaftan u 
and also by Jones and Eyles. 11 The 
high repetition rate of 100,000 pictures/ 
sec is obtained at the rather low aperture 
of ^/64, but the camera is most suitable 
for self-luminous studies in explosion 
research. Each image shows the event 
from a slightly different aspect, and 
this feature must be taken into account 
when reconstructing the spatial develop- 
ment of the rapid event. 

Henry 12 has made and used a simple 
and cheap form of drum camera com- 
bined with repetitive air-spark illumi- 
nation to take excellent pictures of loom 
behavior in cotton processing research. 
The Cotton Research Association is one 
of the groups in England well aware of 
the importance and potentialities of 
high-speed photography. 

In a review of this kind, which at- 
tempts to cover both the past history 
and the present position in this field, 
it is not, perhaps, amiss to describe a 
project still in progress. The author 
has been working in collaboration with 
Barr and Stroud Ltd., for the last four 
years, on an ultra-high-speed drum 
camera in which the images are formed 
rapidly on stationary film by somewhat 
novel means. The camera is designed 
to operate up to 25,000 frames/sec at a 
picture size of 24 mm square. The 
aperture of the system is //8 and all 
images are seen from the same aspect. 
Any standard 35 mm film stock can be 
used in the instrument. 

In drum cameras in which the images 
are formed on the film by a rapidly 
spinning central mirror, the optical 
systems are of two types. In both types 
a primary objective forms an image of 
the event on the axis of the spinning 
mirror. Surrounding this mirror is a 
ring of secondary lenses which form a 
set of images on the stationary film. 
In systems which require image-rotation 
compensation devices, the axis of rotation 
of the spinning mirror is coincident with 



242 



March 1953 Journal of the SMPTE Vol. 60 



the optical axis of the primary objective, 
and the plane of the mirror surface is 
generally at 45 to the primary objective 
axis. Systems which do not require 
image-rotation compensation have the 
axis of rotation of the mirror normal 
to the optical axis of th^ primary 
objective. The Barr and Stroud camera 
is a system of the latter type, and there 
is only one moving part. The central 
prism design is the most important single 
feature of the camera. It consists 
essentially of a sphere of glass divided 
along a diametral plane; the flat 
surfaces are silvered to form plane 
mirrors and the sphere is then joined 
together again and held in a steel cage. 
The prism unit, which is 6 in. in diam- 
eter, spins at 7,500 rpm to sweep 
the images across 100 secondary lenses. 
The film drum is 58 in. in diameter, and 
the whole camera weighs about 1 ton. 
From preliminary photographic tests 
already made the image quality appears 
to be extremely good. A full description 
of the camera will be published on 
completion of the project. An interest- 
ing feature of the design is that the 
instrument can work at unit magnifica- 
tion. 

Light Sources of Short Duration 

Air Sparks. The electric spark in air 
was the earliest light source used for the 
study of rapid events by photography. 
Fox Talbot 13 suggested its use a century 
ago, and the air spark is still used in 
ballistic research laboratories all over 
the world. The time duration is less 
than 1 psec and by the use of delay line 
discharges can be made as low as 0.1 
jusec. Sixty years ago Lord Rayleigh 14 
described the use of the spark for the 
study of air jets and thin films, and at 
the same time Worthington 15 was carry- 
ing out his beautiful work on the study 
of splashes and the impact of liquid drops 
on solid surfaces. Wood 16 studied the 
properties of sound waves in air by these 
methods. Boys, 17 one of our greatest 
experimenters, took the first pictures of 



bullets in flight, and in addition, proved 
by optical means how short was the 
duration of his sparks. Today, Adams 18 
and his team at the Armament Research 
Department use sparks widely for their 
work and have combined this form of 
light source with multiple-gap systems 
for studying the plastic bending of a 
steel rod under very high impact shock. 

Guided sparks, in which the discharge 
is made to take place along prearranged 
channels, permit a more efficient use 
of the light source in combination with 
the optical systems necessary for these 
studies. The relatively poor conversion 
of electrical energy to light can be partly 
overcome by using emulsions having 
high blue sensitivity. 

Repetitive flashing of a single gap is 
also used, the successive images being 
received on a conventional drum camera, 
or on a stationary plate if the transla- 
tional velocity of the event causes the 
images to move across the image plane. 

Sparks in Rare Gases. Condenser dis- 
charges in argon, krypton, xenon and 
other gases have been studied in very 
great detail by H. E. Edgerton. 19 His 
pioneer work, dating from 1931, has 
been reported in this Journal and many 
others. This research has provided a 
strong stimulus to other laboratories, 
particularly where repetitive flashing 
light sources, with a high degree of 
reliability and repeatability of behavior, 
are needed. Energy ranges in the lamps 
have been from one to many thousands 
of joules in the single-flash types, and 
the flash times have ranged from about 
a microsecond to as high as a milli- 
second. In Great Britain Aldington 20 
has made notable contributions to the 
field and developed a wide variety of 
lamps, generally xenon-filled, for many 
research and industrial purposes. 

For ballistic research Mitchell 21 de- 
veloped an argon-filled lamp called the 
Arditron. Filled to high pressure and 
operated from a low-value capacitor, 
the duration can be of the order of a few 



W. D. Chesterman: High-Speed Photography in Great Britain 



243 



microseconds. This tube has been pro- 
duced commercially by several firms and 
is widely used in many laboratories. 

The author and his colleagues have 
studied in some detail the use of xenon 
for very short-duration flashes. 22 The 
work was at first concentrated on making 
a repetitive light source for flashing 
rates up to 4000/sec. Photographic 
duration for a 4-j flash was reduced to 
5 jisec. Stamp and Coghlan 23 have de- 
scribed the method of measurement of 
this duration. The work reached a 
successful conclusion in the development 
of a high-power stroboscope 22 which can 
be used in conjunction with the Acmade 
camera. 4 

The research on xenon is still in 
progress. The effect on the electrical 
discharge characteristics of changes in 
gas pressure, energy in the lamp, and 
tube arc length have been studied. The 
design of the lamp must be integrally 
related to the external circuit conditions 
in the discharge loop. In the author's 
opinion there is still much necessary 
fundamental research to be done on the 
short-duration spark discharge in xenon. 
It is the interrelation of electrical 
transient and photographic afterglow 
which it is most important to under- 
stand. The enhanced light output of 
the xenon discharge over that of the 
air spark makes the former a most 
valuable source for flash work. 

Schlieren Systems 

Schlieren systems have been widely 
used by the Safety in Mines Research 
Laboratory in Great Britain in their 
studies of "fire-damp" explosions. 
Aerodynamic research institutions use 
schlieren optics in various forms for 
airflow research. Among recent in- 
teresting contributions one may mention 
the use by Holden and North 24 of a 
color emulsion for their schlieren images. 
Quantitative analysis is made easier 
by this development. 

The author has used the collimated 
double-mirror schlieren system for the 



study of ultrasonic patterns in a water \ 
tank. The great advantage is that the 
pressure fields are not disturbed by the i 
insertion of any measuring instruments 
or gauges. 

Eleptrooptical Shutters 

There is undoubtedly a wide field of 
application ahead for electrooptical 
shutters, when very short exposure times 
are required. As is well known, the 
simplest system employs the Kerr effect, 
in which rotation of the plane of polari- 
zation of a light beam occurs when a 
high voltage is applied across a glass 
cell containing nitrobenzene. Exposure 
times of 0.1 /xsec have been achieved 
by these systems and Froome 25 has ap- 
plied repetitive pulses at a high rate to 
get a succession of pictures. He was 
studying 26 the behavior of the cathode 
spot on an undisturbed liquid "surface, 
and since the translational movement of 
the self-luminous event was high, suc- 
cessive phases could be studied on a 
stationary plate. The photographic as- 
pects of Kerr cell shutters have been 
critically discussed in some detail by 
Holtham and Prime, 27 members of the 
Liverpool University research team 
under Meek, who have made various 
original contributions in high-speed 
photography over a number of years in 
Great Britain. 

The image-converter tube is now 
beginning to be widely explored as an 
electrooptical shutter of negligible time 
inertia and considerable flexibility of 
application. Courtney-Pratt 28 showed 
the potentialities of the system. In 
addition, Prime and Turnock 29 have 
investigated the use of an iconoscope 
tube and shown that it has promising 
possibilities. 

Some Research Studies 

Space forbids more than a brief men- 
tion of the diverse subjects studied with 
these high-speed methods. In Great 
Britain, as in many countries, some of 
the observational tools of war have 



244 



March 1953 Journal of the SMPTE Vol. 60 



found interesting applications in peace- 
time research. This is particularly true 
in the zoological, biological and medical 
fields. 

Zoological, Biological and Medical Ap- 
plications. Here one may quote motion 
studies of insects, birds and fish in their 
natural environments. Chadwick 30 in- 
vestigated wing-beat rates in his genetic 
studies of Drosophila, and other workers 
have shown experimentally a wide wing- 
beat rate from 6 to 300/sec by these 
means. Hosking 31 has made many 
beautiful studies of birds in flight, using 
flash-discharge illumination. Brown 32 
has studied the wing movement of vari- 
ous birds in the laboratory, regarding the 
movements from the standpoint of a 
mechanical system. The importance 
of different wing muscle groups has been 
clarified. The top speed of the wing of 
a carrier pigeon has been proven by this 
means to be as high as 60 mph. 

The author and his colleagues have 
studied fish movements in slow motion 
with underwater cine cameras operated 
by a frogman with air-breathing self- 
contained equipment. A most interest- 
ing unsolved hydrodynamic problem is 
the motion of a porpoise at full speed, 
which occurs without cavitation. 

In medical research the flash source is 
a most valuable "cool" light. Potter 
and McDonald 33 have studied the speed 
of arterial blood flow. There is un- 
doubtedly much work to be done in 
high-speed photomicrography using flash 
sources. 

Military Applications. The greatest field 
of all applications is of course in military 
science, in particular for ballistic research 
underwater and in air, and for explosion 
research underwater and in air. As in 
U.S.A., much work has been done in 
Great Britain with intermediate-rate 
cameras and flash sources. 

Cavitation research, whether con- 
cerned with propellers, impeller systems 
or hydroelectric schemes, has made 



wide use of high-speed photography. 
The problems of underwater illumina- 
tion have been studied by the author 34 
in relation to many of these problems. 

Recent work which is typical of the 
investigations has been described by 
Senior, 35 and more recently by Taylor, 36 
the latter in respect of one of the atomic 
bomb explosions. 

The Future 

In Great Britain there have been many 
notable contributions to the advance 
of the methods and applications of high- 
speed photography. There is, however, 
still a lack of consciousness, except in 
some major research teams, of how 
great are the potentialities ahead. The 
engineer, the physicist and the worker 
for whom the technique is required must 
form their plans with the greatest co- 
herence. Progress in the future will 
then arise from their common outlook. 

Acknowledgements 

This paper is published by permission 
of the British Admiralty. It is a great 
pleasure to acknowledge the cooperation 
of the many friends and colleagues with 
whom I have worked in the Royal Naval 
Scientific Service on this field of research. 

References 

1. W. D. Chesterman, The Photographic 
Study of Rapid Events, Clarendon Press, 
Oxford, England, 1951. 

2. Anon., "Vinten Type HS. 300, 35mm 
camera." 

3. D. H. Geary, "The Vinten HS. 2000 
Camera," Phot. J., 79: 291, Apr. 1939. 

4. W. D. Chesterman and D. T. Myers, 
"A 35mm high-speed cinematograph 
camera," J. Set. Instr., 28: 301, Oct. 
1951. 

5. W. H. Connell, "The Heape and 
Grylls' Machine for high-speed photog- 
raphy," J. Set. Instr., 4: 82-87, Dec. 
1926. 

6. C. F. Jenkins, "The Jenkins Chrono- 
teine Camera for high-speed motion 
studies," Trans. SMPE, 10: 25-30, 
#25, May 1926. 

7. H. W. Baxter, "A concise report on a 



W. D. Chesterman: High-Speed Photography in Great Britain 



245 



high-speed camera for simultaneous 
photographic and oscillographic rec- 
ords," J. Set. Instr., 79: 183-184, 
Dec. 1942. 

8. F. Brailsford and K. F. Shrubb, 
"High-speed photography of welding 
arcs," J. Sci. Instr., 25: 211-213, June 
1948. 

9. "The Scophony High-Speed Camera," 
Phot. J., 86B: 42-46, #2, Mar. -Apr. 
1946. 

10. Kenneth Shaftan, "Progress in photo- 
graphic instrumentation in 1950," 
Jour. SMPTE, 57: 443-488, Nov. 1951 
(seep. 447). 

11. G. A. Jones and E.- D. Eyles, "Recent 
British equipment and technique for 
high-speed cinematography," Jour. 
SMPE, 53: 502-514, Nov. 1949. 

12. P. S. H. Henry, "A high-speed cine- 
matograph camera," J. Sci. Instr., 21: 
135-141, Aug. 1944. 

1 3. W. F. H. Talbot, Spark Photography, 
Patent No. 13664, June 1851. 

14. Lord Rayleigh, Proc. Roy. Inst., Feb. 
1891. 

15. A. M. Worthington, A Study of Splashes, 
Longmans, New York, 1908. 

16. R. W. Wood, Phil. Mag., 48: 218, 
1899; and 50: 148, 1900. 

17. C. V. Boys, Nature, 47: 415 and 440, 
1893. 

18. G. A. Adams, Unpublished, private 
communication to the author. 

19. H. E. Edgerton, "Electrical-flash 
photography," Jour. SMPE, 52: 8-23, 
Mar. 1949; and numerous other 
articles. 

20. J. N. Aldington, Jour. IEE (London), 
95: Part II, 671, 1948. 

21. J. W. Mitchell, Jour. IES, Dec. 1948. 

22. W. D. Chesterman, D. R. Glegg, G. T. 
Peck and A. S. Meadowcroft, "A new 
power stroboscope for high-speed flash 
photography," Proc. IEE, 98: Pt. 2, 
619, 1951. 

23. W. R. Stamp and R. P. Goghlan, 
"Growth and decay of light measured 



photographically from flash discharge 
tubes," presented on October 9, 1952, 
at the SMPTE Convention at Wash- 
ington, D.C. 

24. D. W. Holden and R. J. North, "A 
colour-schlieren apparatus," Nature, 

, 169: 466, 1952; and Aeroplane, 82: 
' Jan. 4, 1952. 

25. K. D. Froome, "Electrically operated 
Kerr cell shutter," J. Sci. Instr., 25: 
371-373, Nov. 1948. 

26. K. D. Froome, "Behaviour of cathode 
spot on an undisturbed liquid surface 
of low work friction," Proc. Phys. Soc. 
London, 63B: 377-385, June 1950. 

27. A. E. J. Holtham and H. A. Prime, 
"Operation and photographic charac- 
teristics of Kerr cell type electro- 
optical shutter," Proc. Phys. Soc. London, 
63B: 561, 1950. 

28. J. S. Courtney-Pratt, "A new method 
for the photographic study of fast 
transient phenomena," Research 2: No. 
6, 287-294, June 1949. 

29. H. A. Prime and R. C. Turnock, 
"Iconoscope electrooptical shutter for 
high-speed photography," Rev. Sci. 
Instr., 20: 830, 1949. 

30. L. E. Chadwick, Physiological Zoology, 
1939. 

31. E. Hosking, Country Life, Nov. 5, 1948. 

32. R. H. J. Brown, Jour. Exp. Biology, 
25: 322, 1948. 

33. J. M. Potter and D. A. McDonald, 
"Cinematographic recording of velocity 
of arterial blood flow," Nature, 166: 
596-597, Oct. 1950. 

34. W. D. Chesterman and J. B. Collins, 
"Problems of underwater illumina- 
tion," Trans. IES (London}, 17: 193, 
1952. 

35. D. A. Senior, "High-speed photog- 
raphy of underwater explosions," 
Phot. J., 86B: 25-31, Jan.-Feb., 1946. 

36. G. I. Taylor, "The formation of a blast 
wave by a very intense explosion," 
Proc. Roy. Soc. (London), A201: 175-186, 
Mar. 1950. 



246 



March 1953 Journal of the SMPTE VoL 60 



Rapid-Sequence Camera 
Using 70mm Film 

By CHARLES A. HULGHER 



This paper presents the problem of obtaining relatively large photographs 
in sequence at frame rates up to 50/sec. Design characteristics required for 
a camera suitable for obtaining photographic data of missile launching* and 
flights with negatives 2j in. wide and 2^ or 5 in. long are enumerated. The 
intermittent film transporting mechanism and shutter system used in the 
solution of this problem are described. 



JL HE ADVANTAGES of a large negative 
for detail study of photographic data 
are thoroughly recognized and are 
employed wherever the existing condi- 
tions permit. However, in cases where 
photographic data are required in se- 
quences of more than several frames 
per second, the desirable characteristics 
of the large negative have, of necessity, 
been compromised in favor of the 35mm 
or 16mm motion picture frame sizes. 
The Langley Aeronautical Laboratory 
of the National Advisory Committee for 
Aeronautics at Langley Air Force Base, 
Virginia, has developed a rather unique 
camera capable of producing relatively 
large negatives at sequence rates con- 
siderably higher than heretofore avail- 
able. The basic principles of this camera 
have since been incorporated in a re- 
fined model which has recently been 
placed on the market. While this 

Presented on October 10, 1952, at the 
Society's Convention at Washington, D.C., 
by Charles A. Hulcher, Charles A. Hulcher 
Co., Inc., 40 Manteo Ave., Hampton, Va. 



camera was developed primarily to 
obtain photographic data from rocket- 
powered missiles, it has proven especially 
useful in numerous other fields. These 
instruments are now in service in several 
stations in this country and in Canada. 

The Problem 

Preliminary specifications were estab- 
lished as follows: 

(1) Provide a negative size at least 
2 in. wide and 5 in. long. This was 
considered a minimum frame size for 
fixed cameras photographing missile 
launchings. The 5-in. dimension could 
be aligned with the proposed missile 
trajectory. This would provide for 
the use of focal lengths long enough to 
obtain the required image sizes and still 
maintain enough field to insure photo- 
graphic coverage. 

(2) Provide minimum sequence speeds 
of at least 10 frames/sec. 

(3) Provide a shutter system which 
would efficiently stop the motion without 
excessive distortion. 



March 1953 Journal of the SMPTE Vol. 60 



247 




Fig. 1. Exterior view of the Hulcher 70mm rapid-sequence camera. 




Fig. 2. Interior view of the film transporting system; 
see text for description of keyed parts. 



(4) Provide for a high degree of 
reliability. This was considered very 
important since jamming or other 
malfunctioning of the camera could 
result in total loss of records in the event 
of some mishap at launching. While 
these specifications covered the major 
requirements, it was also considered 
desirable to develop a camera rugged 
enough to withstand the abuse asso- 
ciated with the installation of these 



instruments on a sandy ocean beach, 
be reasonably simple to operate and 
maintain, utilize the available lens 
systems of the K24 Aero Camera and 
not be excessive in weight or cost. 

The Solution 

A prototype camera was subsequently 
designed and built and has since proven 
successful in meeting the requirements 
set forth in the specifications. This 



248 



March 1953 Journal of the SMPTE Vol. 60 



camera utilized 70mm perforated film 
on 100-ft daylight-loading spools. It 
produces negative sizes of either 2\ in. 
by 5 in. or 2\ in. by 2\ in. It operates 
at variable sequence speeds up to 
twenty-five 5-in. pictures per second 
or up to fifty 2^-in. pictures per second. 
Its double rotating shutters, operating 
close to the focal plane, provide high 
motion arresting characteristics. The 
camera contains no reciprocating parts 
and since all moving parts are balanced 
and rotate at a constant preset speed, 
the camera has proven remarkably 
reliable and trouble-free. Rugged cast 
aluminum alloy construction and the 
absence of any small or intricate parts 
have proven more than equal to the 
abuses inherently associated with this 
type of installation. 

Description 

The exterior view shown in Fig. 1 
indicates the general form of the camera. 
The lens mount slides in machined 
grooves in the front casting and may be 
removed to insert the desired lens. 
Focusing is accomplished by rotating 
the knob on the top of this housing which 
swings a front surface mirror into a 
45 position reflecting the image onto 
the ground-glass screen for focusing. 
The 6-in. square base is designed to fit 
standard tripods. The camera weighs 
approximately 30 Ib. Figure 2 is an 
interior view of the camera's film trans- 
porting system. Film supply from 
spool D is fed over the sprocket and 
held in position by the film hold-down 
device G. It then passes in an easy 
loop over roller I through the pressure 
plate behind roller G, under the inter- 
rupter E and thence under the sprocket 
and on to the take-up spool. Both the 
sprocket and the interrupter rotate in 
a counterclockwise direction and at the 
same speed. Each revolution of the 
sprocket feeds 5 T ^ in. of film in on the 
topside and out on the bottom to the 
take-up spool. Except for the action 
of the interrupter, the film runs at a 



constant speed from the supply spool 
through the camera and on to the take- 
up spool. Rotation of the interrupter, 
however, changes the length of the film 
path between rollers G and F. This 
lengthening of the film path causes 
the loop above the pressure plate to be 
decreased when the film path below 
the pressure plate is extended, thus 
accelerating the film through the pres- 
sure plate. As the interrupter rotates, 
however, it leaves the film, causing an 
excessive loop to be formed below the 
pressure plate. While this loop is 
being consumed, there is no movement 
of the 5-in. strip of film held firmly by 
the constant force exerted on it by the 
pressure plate. The rotating-disk shut- 
ters are geared with the sprocket and 
interrupter in such a manner as to 
expose the film completely before it is 
again placed in motion. The double- 
roller arrangement on the interrupter 
causes the film in the pressure plate to 
stop twice for each 5^ in. of film 
supplied by the sprocket. The shutters, 
however, are so arranged as to expose 
the film only once for each complete 
rotation of the sprocket and interrupter 
when taking 5-in. pictures. This shutter 
is replaced by another having double 
openings, and a mask is inserted in the 
aperture for taking 2j-in. pictures. In 
this manner the frame size is reduced 
from 5 in. to 2j in. in length and the 
camera produces pictures at twice the 
5-in. frame rate while rotating at the 
same speed. 

The shutter system is composed of 
two 9^-in. diameter disks, Both of which 
rotate in the same direction. The 
shutter closest to the film plane serves 
to expose the film and is geared to make 
four complete revolutions for each 
5^ in. of film fed by the sprocket. This 
provides a high traversing speed re- 
sulting in a short time increment be- 
tween the top and the bottom of the 
5-in. aperture. This exposing disk has 
one 21 opening. To prevent the film's 
being exposed four times, another 



C. A. Hulcher: 70mm Sequence Camera 



249 





Fig. 3. Contact prints of a sequence taken at 45 frames sec. 



250 



March 1953 Journal of the SMPTE Vol. 60 




C. A. Hulchcr: 70mm Sequence Camera 



251 



shutter having one 90 opening rotates 
directly in front of the exposing shutter 
and at one-fourth the speed of the 
exposing shutter. This masking shutter, 
therefore, prevents film exposure for all 
but one of the four revolutions made by 
the exposing shutter. Both shutters are 
aligned with the interrupter in such a 
manner as to provide exposure of the 
5-in. film strip only while the film is 
motionless. The exposing shutter was 
purposely designed with a fixed opening 
in the interest of simplicity and to lessen 
the chances of malfunctioning which may 
arise with an adjustable opening. Since 
the exposing shutter, however, consists 
primarily of a simple flat disk, other 
disks with different openings may easily 
be substituted. Although the size and 
arrangement of the interrupter have 
been designed to stop the film for a 
minimum interval, it is, nevertheless, 
possible to increase the opening in the 
exposing shutter up to 70 and still 
obtain exposures before the film is 
again in motion. Exposure time in 
seconds is calculated as follows: 

Shutter opening degrees 



360 



5-in. frames /sec X 4 

Exposure time in seconds. 

To change from 5-in. to 2|-in. pic- 
tures, the masking shutter with one 
90 opening is removed and replaced 
by a similar shutter having two dia- 
metrically opposite 90 openings. This 
masking shutter is keyed to the shaft in 
the same position as the 5-in. frame 
masking shutter. Since -the interrupter 
stops the film twice for each 5 ^o i n - of 
film supplied for each revolution, the 
double opening shutter allows the ex- 
posing shutter to make exposures twice 
for each revolution of the interrupter. 
A mask having an opening 2^ in. high 
is slipped into the grooves provided in 
the aperture. In this manner, 2^-in. 
pictures are provided at twice the se- 
quence speed of the 5-in. pictures with- 



out increasing the speed of the camera. 
Inasmuch as the exposing shutter is 
rotating at the same speed, the film 
exposure remains the same for the 2j-in. 
frame size as it was for the 5-in. frame 
size, even though the camera is taking 
pictures twice as fast. The 5-in. pic- 
tures are spaced approximately 0.1 in. 
apart and the 2^-in. pictures approxi- 
mately 0.05 in. apart, thereby resulting 
in a minimum film waste. 

Lenses are mounted on a lens plate 
which slides in machined grooves on 
the front of the shutter housing. This 
housing is hinged and may be swung 
open to change shutters. When closed, 
it is locked firmly against machined 
surfaces and aligned by a conical ring. 
Lenses of from 5-in. to 40-in. focal 
lengths have been utilized and several 
installations are considering the use of 
considerably longer focal lengths. The 
2j in. by 2j in. frame size provides 1\ 
times the area of the 35mm standard 
frame. It is possible, therefore, to track 
missiles with the same accuracy as 
obtained with the 35mm camera with 
lenses of more than twice the focal 
length employed with 35mm cameras. 

This camera is powered by a f-hp 
series motor with a variable-speed 
centrifugal electric governor. Its roomy 
interior provides adequate space for 
special timing devices, either within the 
lens mount when the timing system 
requires the use of the shutters, or at or 
near the sprocket, when the timing 
system requires that the film be traveling 
at a constant rate. When timing is 
inserted on the lower side of the sprocket, 
the film distance between the timing 
marks and the exposure remains con- 
stant irrespective of the size of the loop 
used when threading the camera, since 
the camera automatically fixes the lower 
loop. 

It is believed that this camera will do 
much to fill the gap which has so long 
existed between the small frame size 
of the movie camera and the large 
frame size of the still camera. 



252 



March 1953 Journal of the SMPTE Vol.60 



Precision Film Editor Utilizing 
Nonintermittent Projection 



By TORBEN JOHNKE 



Maximum efficiency in editing film both in production studios and in television 
stations is the objective of a recently developed preview-and-editing machine 
which utilizes a continuous projection process in place of the customary inter- 
mittent movement and shutter. Because projection is continuous, and because 
all parts are so machined that no portion of the film except the sprocket-hole 
area ever comes into contact with any surface, valuable originals and fine-grain 
masters can safely be edited on this machine. 



NE IMPORTANT WAY to reduce COStS of 

production is to minimize time lost in 
previewing and editing. To attain 
utmost economy in those two processes 
three requirements must be fulfilled. 

First, the equipment used must be of 
such nature that valuable originals and 
masters can be entrusted to it without risk 
of damage. Otherwise editing and pre- 
viewing must be deferred until after 
duplicates have been made. If produc- 
tion must be suspended pending pre- 
viewing, the time thus lost can be of 
significant importance. 

Second, the equipment used should be 
sufficiently versatile in design and details 
to meet all normal demands of studio and 
telecasting procedures without any delay 
for modification of facilities or special 
setups of any kind. 



Presented on April 25, 1952, at the Society's 
Convention at Chicago, for R. M. Savin i, 
by Torben Johnke, Editor Precision Equip- 
ment Corp., 130 W. 46 St., New York 36, 

N.Y. 



Third, the equipment should permit 
several persons to see the picture and 
hear the sound simultaneously. And if 
the picture is bright enough for several 
persons to see it clearly under ordinary 
room lighting that is a further advantage. 

The Precision Film Editor is a device 
engineered in every detail to meet the 
above requirements completely and with 
generous tolerances. It is versatile and 
flexible in operation to a degree that is 
unprecedented in operation ; if used with 
only ordinary care, even by unskilled 
personnel, it can be trusted absolutely 
not to damage film; it presents a 7 X 9 
in. screen image bright enough to be seen 
clearly under ordinary room illumination 
and a loudspeaker output of 4 w. 

Figure 1 is a general view of the ma- 
chine. At top, rear, is the screen, flanked 
on one side by the loudspeaker grille and 
on the other by a dummy grille included 
for, symmetry of appearance, behind 
which is located the amplifier. 

The projector mechanism and sound- 



March 1953 Journal of the SMPTE Vol. 60 



253 




Fig. 1. The Precision Film Editor. 





I 



254 



Fig. 2. Panel control board. 
March 1953 Journal of the SMPTE VoL 60 



ead are grouped on the panel in the 
enter of the table, with the projector 
earer the screen and the soundhead at 
tie rear, nearer the operator's position. 
The two turntables at left are intended 
or picture and sound reels, respectively, 
he one nearer the screen carrying the 
>icture film and the one nearer the opera- 
or, the sound film. The corresponding 
ake-up turntables are at the right, 
lowever, right and left can be reversed, 
ince motor direction can be reversed 

th ease. 

The controls are seen, conveniently 
ocated for the right hand, on the front of 
he desk. The chromium control at left 
s for adjusting synchronism between pic- 
ure and sound. The sound reel always 
uns at constant speed although that 
peed may be varied; the chromium 
ontrol is used to adjust the speed of the 
Dicture reel faster or slower with refer- 
ence to the sound reel, until synchronism 
lis attained. A frame counter shows 
pxactly how many frames displacement 
Were needed to achieve synchronism. 
At the right of the synchronization con- 
jtrol wheel is the speed control knob. 
This alters the speed of both sound and 
picture reels simultaneously between 
imits of 2 and 56 frames/sec. To the 
right of the speed control is a row of 
switches for the amplifier, motor, etc., and 
a volume control. The foot pedals can be 
used instead of the hand control, if pre- 
ferred, to set the equipment into opera- 
tion. Pressure on the right pedal drives 
the films forward; pressure on the left 
pedal automatically disengages the for- 
ward drive and throws in reverse drive. 

Figure 2 shows the panel controls in 
greater detail, and also shows the editing 
machine with separate picture film and 
sound film threaded in, in one of the 
three ways of threading. 

There are two other ways of threading. 
When previewing or editing combined 
picture and sound prints, the picture film 
is threaded through the aperture and 
then run directly to the soundhead with- 
out loops. Since the projection process 



is continuous and an intermittent move- 
ment does not exist in this mechanism, no 
loops are needed; and if the film is 
threaded taut the displacement between 
picture and sound apertures is always 
exactly 19 frames. When separate 
picture and sound prints are used, and 
the purpose is to edit dialogue or per- 
form any other function for which sound 
flutter is not important, the sound film 
can be threaded through one sprocket 
only instead of two for faster, easier 
operation. 

There are two general methods of 
editing. In one, the editor uses the sur- 
face of the editing machine as a rewind 
bench, and cuts and splices the film as he 
proceeds. In the other, the editor makes 
notes which are passed on to another 
person who subsequently does the cutting 
and splicing. These notes may include 
information on the number of frames dis- 
placement needed for synchronism. They 
may also include references to the film 
markings made by the editor, for this 
machine has two film markers, one at the 
picture aperture and the other at the 
sound aperture. Figure 3 shows the 
latter in use. A small chip is nicked out 
of the film. These chips drop into con- 
tainers provided for them, which must 
be emptied occasionally; loose chips 
might conceivably get into the drive and 
scratch film, hence these special con- 
tainers which make any such occurrence 
impossible are provided. 

The drawers at the right and left side 
of the machine provide reasonable stor- 
age space for film, reels, splicing ma- 
terials, note pads and other facilities for 
efficient operation. 

Figure 4 shows some of the mechanical 
details of the drive. The motor is 
wound for 220 v, 3-phase a-c, and has a 
built-in stepless gear for continuous varia- 
tion of driving speed. In Fig. 4 we are 
looking at the machine as it would 
appear if laid over on its back ; the under- 
side of it faces us and the control panel is at 
the top. The chromium synchronization- 
control wheel is just visible at the top, 



Torben Johnke: Film Editor 



255 




Fig. 3. Film marker at the sound aperture. 




Fig. 4. Underside of machine's table, showing drive; control panel at top. 
256 March 1953 Journal of the SMPTE Vol. 60 




Fig. 5. Open rolls of film on machine. 

toward the right; from it a shaft with a 
universal joint extends diagonally down- 
ward. Just right of the synchronization- 
control wheel is the speed control knob; 
from it a shaft extends vertically down- 
ward to a sprocket; and from this 
sprocket a chain runs leftward to the 
housing of the combined motor and step- 
less speed control gear. The V-belt 
drives to the four turntables are plainly 
visible, and at the extreme left there is a 
glimpse of one of the relays that control 
and reverse the drive motor. 

The machine can accommodate either 
open rolls of film, or film on reels, as 
shown in Figs. 5 and 6 ; there is no need 
to take time out and delay proceedings 
merely to put film on a reel or take it off 
one it can be taken either way it comes. 
How this is accomplished is shown in 
Fig. 7, where the spindle for open rolls 
has been lifted off and placed to one 
side, revealing the spindle used when film 
is on a reel. Up to 2,000 ft of film in one 
roll or reel can be accommodated. 

The heart of the Precision Film Editor 
is the shutterless continuous projector. 

The lamp used for projection light is 
actually a sound exciter lamp of 10 v, 
?i amp rating. Screen illumination at 
75 w is ample. 

From the lamp the light path is 
through a condenser lens and then through 
the film. Film motion is, as the pictures 
show, lateral instead of vertical. After 
passing through the film the light enters 
and traverses a 12-side revolving prism. 
This prism can be described as a disk of 
optical glass as high as a frame is wide, 
rotating horizontally behind the film; 



Fig. 6. Film on reel on machine. 




Fig. 7. Spindle for open rolls placed to 
one side to show spindle for film on reels. 

this disk, however, is not circular, for its 
rim has been ground to a duodecagon. 
Each prism face has approximately the 
same dimensions as one frame; and the 
rate of rotation is 2 prism faces per frame 
in the same direction as the film. 

On emerging from this rotating 1 2-face 
prism the light beam passes through the 
projection lens and then into an assem- 
blage of three motionless triangular 
prisms. These latter perform three func- 
tions: by their agency the light beam is 
deflected upward; the image is erected 
so it will appear erect on the screen; 
and, finally, the light beam is deflected 
forward, above the top of the rotating 
prism housing, to the screen. 

In Fig. 2 the projection lamp and con- 
denser lens are inside the housing nearest 
the screen. Thence the light path is 
forward, through the film, toward the 
operator. Having passed through the 
film, the light proceeds to and through 
the rotating prism under the circular 
housing. On the near or left side of the 



Torben John ke: Film Editor 



257 




Fig. 8. Film editor in use, showing synchronization control. 



circular housing can be seen the turret 
arrangement within which are the pro- 
jection lens and three stationary, tri- 
angular prisms. From the top of this 
little turret the light beam is projected 
straight across the prism wheel and 
lamphouse to the screen. 

In operation, this mechanism produces 
a screen image that is always in frame. 
The image is free from flicker even at the 
low projection speed of 2 frames/sec; 
successive frames appear to merge into 
one another. This of course is what 
they do, since there is no shutter. 

This nonintermittent projection sys- 
tem needs no gate and has none. At no 
point in the film path does the picture 
area come into contact with any me- 
chanical surface. Operation is vibration- 
less. At speeds up to 24 frames/sec 
operation is absolutely silent; at higher 
speeds some machine noise may become 
apparent. 



Because projection is nonintermittent 
the film can be brought to full speed from 
a standing start within one frame with- 
out danger of damage; and it can be 
stopped and reversed in direction \vith 
similar rapidity without danger of dam- 
age. With intermittent projection, or 
with a projection gate, this would be 
impossible. 

The soundhead in Fig. 2 is the panel 
assemblage closest to the operator. The 
curving hood at the front of the panel 
houses the exciting lamp. Since the 
film motion is lateral, this is a lamp with a 
vertically-disposed filament. It is of 6-v, 
1-amp rating, and powered by a super- 
audible a-c frequency. The photoelec- 
tric cell is inside the drum just to the rear 
(right) of the exciting lamphousing. 

The comfort of the editor or other 
operator has been fully considered. The 
foot pedals of Fig. 1 are cable-connected, 
and enough slack in the cable has been 



258 



March 1953 Journal of the SMPTE Vol. 60 



provided so they can be drawn forward, 
out from under the machine. For ex- 
ample, the continuity girl might put her 
typewriter alongside the film editor and 
these foot pedals under her typing table. 
Figure 8 shows the film editor in use. 
The operator is relaxing comfortably, 
with his right hand on the synchroniza- 
tion control. The resynchronizing frame 
counter can be seen clearly just to right 
of the exciting lamp hood ; on the other 
side of that same hood is the footage 
counter and a continuous frame counter. 



Footage and frame counters can be 
reset manually to zero at any time. 

The Precision Film Editor is available 
for 16mm as well as 35mm film, and also 
for 1 7mm ; but not as yet for -in. tape. 
A magnetic soundhead has been de- 
signed, and can be added to the equip- 
ment shown in these pictures. Provision 
has also been made for using headphones, 
when desired, in place of the loudspeaker 
so the equipment can be used in a noisy 
location without disturbance to others or 
distraction to the user. 






Torben Johnke: Film Editor 



259 



The Bridgamatic Developing Machine 



By JOSEPH A. TANNEY and EDWARD B. KRAUSE 



A need was felt for a reasonably priced, quickly accessible, self-contained 
automatic film processor for television stations and small laboratories. This 
led to the design of the Bridgamatic machine which embodies standard com- 
mercial design plus a continuous overdrive, tension-relieving devices, straight- 
line film flow and ease of handling. 16mm and 35mm negative-positive and 
reversal models are described. The add-a-unit idea was adopted, so the bare 
machine can later be equipped with whatever refinements are desired. 



J_ HE GOAL OF SMALL FILM PRODUCTION 

units has long been quick processing. It 
has been attained by some, like the 
big-city television stations which telecast 
"spot" newsreels of the day's events the 
same night. Many racetracks, too, 
project an entire event within a few 
minutes after the finish. But the equip- 
ment cost has been almost prohibitive 
for educational institutions, small pro- 
ducers and commercial film studios 
catering to the lesser television stations. 
The Bridgamatic was designed to fill 
this need for a compact, self-contained 
automatic processor. After several years 
of experiments, a practical, simple 
drive mechanism was designed which 
would be reasonably tamperproof and 
require no special skills to operate. It 
was planned that with intelligent han- 
dling it should keep the film moving for 
hours without much operator attention, 



Presented on October 8, 1952, at the 
Society's Convention at Washington, D.C., 
by Joseph A. Tanney, who read the paper, 
and Edward B. Krause, S.O.S. Cinema 
Supply Corp., 602 W. 52 St., New York 19. 



other than changing reels. Exposed 
film was to be fed into one end so it 
would emerge at the other end, de- 
veloped, fixed, washed, dried and 
reeled ready for projection or printing. 
This seemed at first a comparatively 
easy problem; but because film stretches 
when it is wet and again contracts 
when it dries there is a constant and 
continuous variation in its linearity. 
It was found that a fixed speed drive 
would not take care of this without 
manual readjustment while the machine 
is running. So, the overdrive system 
was adopted. 

Function of the Overdrive 

This equipment operates through the 
medium of two separate drive systems 
combined to produce an overdrive on 
the upper banks of rollers (Fig. 1), thus 
eliminating any troublesome drive com- 
ponents immersed in the solutions. The 
basic drive consists of a chain which is 
driven through a conventional gearbox 
and a drive motor. This in turn drives 
a series of chain sprockets synchro- 



260 



March 1953 Journal of the SMPTE Vol. 60 





T,rrr- 




Fig. 1. Bridgamatic drive compartment with cover removed, showing individually 
driven phenolic yieldable clutch pulleys. 



nously, one sprocket for each bank of 
rollers in the machine. Permanently 
affixed to each chain sprocket is a 
laminated bakelite phenolic pulley of 
the same diameter. This serves as a 
connecting medium to the secondary 
drive system, thus providing the means 
of operating the upper banks of lami- 
nated phenolic bakelite pulleys. A 
belt connects the upper and lower 
pulleys, thus making the lower syn- 
chronous pulley the actuating medium 
and the upper pulley the driven medium. 
Attached to each upper sheave is a 
yieldable friction clutch. These upper 
driven pulleys decrease in diameter 
from the feed-in end of the machine 
through the drybox to the take-up. This 
tapering diameter sheave system in- 
creases the peripheral speed of, and 
induces an overdrive in, the upper 
banks of rollers, each bank rotating at 
a greater rate than the one behind it, 



and so on throughout the entire ma- 
chine. The yieldable friction clutches 
hold the overdrive in check, and will 
not allow any unnecessary strain or pull 
on the film strands at any time. Each 
clutch is adjustable as to tension by a 
tempered compression spring held by a 
locking collar. Significantly enough, 
these springs in themselves contribute 
little to produce the overdrive in the 
Bridgamatic machine. 

This overdrive is mathematically cal- 
culated when the pulleys are made 
the amount of overdrive is positive for 
the life of the machine and has no 
dependency at all on critical spring 
pressures or other fine adjustments. 
Through this means of transmission it 
is also possible to vary the amount of 
overdrive at any one section of the 
machine. For instance, in the ex- 
tremely long machines used for color 
processing, it is possible to eliminate 



Tanney and Krause: Developing Machine 



261 




Fig. 2. Operator inspecting film before making quick splice while 
machine is still running. 



completely the overdriving tendency in 
the drybox where the film is contracting 
while it is drying. In the usual types 
of Bridgamatic machines for small 
laboratory use, the overdrive is carried 
through from the film entry point right 
to the take-up reel. Even the film in 
the drybox has an overdrive working 
there which is also controlled by the 
yieldable clutches. The springs on the 
clutches are merely sufficient to place 
the upper banks of rollers in operation 
and the overdrive automatically takes 
over from there. This action takes 
place continuously in all sections of the 
machine, even though the movement is 
normally imperceptible. 



Adjusting the Overdrive 

Each individual bank or group of 
rollers is adjustable to tension by re- 
leasing the locking collar setscrew using 
an Allen wrench and compressing or 
releasing the spring as necessary. An 
overall adjustment can be made quite 
easily and simply. The film is hand 
held at the feed-in end and the machine 
is started. Each bank of rollers will 
in turn automatically stop rotating. 
The film is then released and the rollers 
are observed as they resume driving. 
Each bank should start successively 
without hesitation the moment the 
tension is released. The first should 
start, then the second, and so on, until 



262 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 3. Interior of stainless-steel tanks showing nylon bearings, stainless-steel 
shafts and other parts of inert materials. 



the take-up reel is turning and the entire 

I machine is in operation. If there is any 

; lagging, the cycle is repeated to locate 

! the offending bank, then the lock collar 

i is loosened and more spring tension 

applied as necessary to that particular 

section. 

When this drive system was being 
designed one of the requirements which 
seemed important was to keep the drive 
mechanism completely clear of any 
chemical contact with solutions. An- 
other important item was to keep the 
machine as wide open as possible and 
free of complicated covers and in- 
accessible places. This was solved by 
enclosing the entire drive mechanism in 
a separate housing of its own (Fig. 1) 
where it is completely out of sight and 
away from solutions, tanks, etc. The 
entire film roller assembly can be 
removed from the machine in a half 
hour with nothing more than an Allen 
wrench. This provides for simple 
periodic cleaning of the tanks and rollers 
if necessary without a total breakdown 
of the machine with its attendant layup 
period. 



Control of Film Transport 

Tests made at the factory with this 
type of drive showed that it was possible 
to run single 8mm film through the 
machine without any film breakage. 
A splice using only one staple did not 
let go throughout the entire process. 
By alleviating the stresses and strains, 
this machine's drive exerts so little 
driving torque that it produces a very 
clean film product and reduces abrasion 
marks on the base side of the film to an 
absolute minimum. Test strips (or any 
particular section of the work going 
through) may easily be removed at 
any time, in any stage of processing. 
This is done by simply grasping both 
ends of the film to be removed, making a 
quick cut and stapling the loose ends 
(Fig. 2). 

Controls and Connections 

One master switch controls the entire 
assembly. Thus one cannot run the 
machine without the heater and blower 
in the drybox or the refrigeration system 
also being in operation. This prevents 



Tanney and Krause; Developing Machine 



263 



film from coming out of the machine 
wet. 

Heliarc- welded 18-8 stainless-steel in- 
sert tanks are regularly supplied to- 
gether with stainless-steel film guides or 
grids and simplified lifts at the bottom 
of each tank. These keep the film on 
the lower banks of rollers when lifted for 
threading or cleaning the phenolic 
bearings. All parts in contact with 
solutions are of stainless steel or other 
inert materials (Fig. 3). 

This equipment requires a minimum 
of plumbing. In the absence of per- 
manent connections, a garden hose 
fits the two pipe fittings protruding 
from the wash tank. The lower pipe 
connects to the darkroom water supply 
and the upper pipe to the dram or 
sewer. This gives a circulating water 
supply in the wash tank. If a rapid 
water change is desired, a portable 
drain pump is useful in several ways. 
It will pump all solutions into and out 
of the tanks and act as a draw-off pump 
on the upper wash drain. A smaller 
pump, especially for overflow draw-off, 
is also available as optional equipment. 

There are three speed-change pulleys 
mounted on the drive motor at the 
end of the standard machine: small 
for negative processing, intermediate 
for sound-track processing and large 
for positive film processing. A recently 
announced optional feature is a 10 to 1 
ratio variable speed control trans- 
mission which permits an unlimited 
range of developing times from 2 min 
to 20 min. This entirely enclosed 
assembly is integral with the main 
drive motor, and replaces the step 
pulleys with their three fixed speeds. 
A variety of formulas can, of course, 
be used and the raw stock suppliers 
can provide suitable developing and 
fixing mixtures. Additional formulas 
can also be found in the Photo Lab 
Index.* The average developing times 



* Photo Lab Index, Morgan and Lester, 
Publishers, 101 Park Ave., New York 17. 



are 6 to 7 min for negative, 5 min for 
sound track and 2 to 2f min for 
positive film. Raising solution tempera- 
tures to an average of 90 F decreases 
the developing-time cycle, and increases 
the machine's output considerably. An 
immersion heater for the water jacket 
with thermostatic control and an extra 
Calrod strip heater in the drybox can 
be installed as original equipment, or 
added later. 

A special Bridgamatic high-speed Re- 
versal 16mm machine to process at 100 
fpm is also available. It is designed for 
ultra linear operation with but 100 ft of 
leader in the machine and it allows 200 
ft of picture to be completely processed ] 
in 3 min. It operates at 125 F using a 
pressurized spray and the latest prin- 
ciples embodying rapid drum-type dry- 
ing. The machine can be designed to s 
exit the film head or tail first, which- ] 
ever is desired. This unit is entirely ' 
automatic, including thermostats, pumps, 
filters and by-pass controls. 

Savings Effected in Chemicals 

The machine will operate with only 
one to two gallons of developer and 
produce excellent results, effecting a 
considerable saving in chemicals on 
small jobs. Normally for larger runs, 
quantities of 2 J to 5 gal are used. There 
is a replenishment formula furnished 
with most developers and over long 
periods this should be added as in- 
structed. Ordinary hypo formulas with 
hardener added are used in the third 
or fixing tank. Air squeegees are 
provided for blowing the water from 
the film before it enters the drying 
cabinet. It is necessary, of course, to 
have a compressor of proper capacity 
operating the squeegees. This must 
be the type that puts out oil-free air, 
although a filter is recommended as 
well. 

Color Processors Also Available 

The basic processing machine is 
of a negative-positive design with four 



264 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 4. Bridgamatic TV Reversal Model showing daylight magazine, 
feed-in elevator and light-tight cover for first three tanks. 



anks: developer, rinse, fix and wash. 
Vficrofilm models have an added wash 
ank with spraybars. A length of 
rom 200 to 400 ft of leader is required, 
md tank capacity varies from 2\ to 
1 5 gal, depending on the model, whether 
16mm, 16/35mm combination or 16mm 
eversal. 

The most popular model at present is 
:nown as the Bridgamatic TV Reversal 
Special, a nine-tank machine measuring 
} ft long, 2 ft wide and 4 ft high, weigh- 
ng 650 Ib. It features a daylight 
oading magazine, feed-in elevator with 
ising indicator and buzzer alarm, 
ake-up elevator, re-exposure lamp and 
wo heats in the drybox (Fig. 4). Aera- 
ion or bubble agitation with a con- 



trollable valve is provided for the 
bleach tank, with a spray header in 
the final wash. Speed is rated con- 
servatively at 750 ft/hr. The new 
High Speed Rapid Reversal pan films 
recently introduced (such as Du Font's 
931) have increased this speed ap- 
proximately 50%. The machine is 
made for operation under average room 
lighting and consumes but 20 amp. 

These machines are now also custom 
built to a variety of speeds and sizes, 
from 16mm to 70mm. Two new 
models are being built for the new 
Eastman Negative-Positive Color Process 
or Ansco Negative-Positive Color Proc- 
ess. It may soon no longer be neces- 
sary for the film producer to think of 



Tanney and Krause: Developing Machine 



265 




Fig. 5. Top view of Bridgamatic 16/35mm machine showing 
straight-line design with all major parts instantly accessible. 



color in terms of outside laboratory 
work and expense. Heretofore color 
processing machines were huge, costly 
and cumbersome, and certainly only an 
investment for the largest commercial 
laboratories since the cost of chemicals 
alone was tremendous. This small, 
reasonably priced color processor is 
expected to save many times its cost in 
a short time as the most expedient way 
of processing and controlling color 
quality on the spot, without the delays 
and accelerated production costs which 
go with expensive retakes. 

Add-a-Unit Construction 

The basic Bridgamatic machine of 
any model with its bare essentials may 



be installed first. Certain attachments 
are required in all installations such as: 
an air compressor to activate air squee- 
gees and supply aeration; a drain 
pump for changing solutions and clean- 
ing out the tanks; air and water filters; 
and a film stapler with rust-resistant 
staples. Other recommended attach- 
ments are: recirculating pumps and 
filters; draw-off pumps; time-delay 
feed-in or take-up elevators; refrigera- 
tion with temperature controls; daylight 
loading magazine; sub-base to house 
accessories and to raise the machine to a 
more convenient working height; re- 
plenishment system; hydraulic speed- 
control system; and top overflow and 
bottom drain valves. All may be 



266 



March 1953 Journal of the SMPTE Vol. 60 



urchased as original equipment or some 
nits may be added later. Permanently 
ffixed units which involve plumbing 
lould be incorporated with the original 
lachine. 

The advantages of this add-a-unit 
rrangement are considerable, for the 
lachine's usefulness can be expanded as 
ic needs arise. Budgets often arbi- 
arily limit the amount which can be 
pent at one time, so breaking up the 
quipment into smaller components 
roves a distinct advantage. 

iase of Servicing 

Important parts are instantly and 
asily accessible (Fig. 5). A complete 
nockdown service job can be done in a 

w hours right on the premises where 
ne machine is installed. Major re- 

lacements can be made quickly, as 
lost assemblies are standard and inter- 
ihangeable. There are no expensive 
jigging or erection costs as the machine 
s shipped ready to plug in and operate. 
Extreme portability and comparative 
ight weight enhance the utility of the 
iipparatus, while the cost makes it a 
practical equipment for even the smallest 
rganization seeking automatic film 
Drocessing equipment. 

Discussion 

Bernard L. Elias (Eastman Kodak Co.) : 
v what method or channels are the re- 



circulated solutions dispersed within the 
tanks? 

Mr. Krause: What we use is a stainless- 
steel recirculating pump and the developer 
is drawn down through the input end of the 
pump and then is pumped back in through 
a stainless-steel line to the tank and through 
stainless-steel barrels which have various 
size holes depending upon the solution 
tanks and the size of the pumps and we use 
impingement jets of solution below the 
surface of the solution itself. 

Ralph Lovell (NBC, Hollywood): With 
this type of overdrive mechanism is it 
not true that if the film stops in, let's say, 
the drybox, or somewhere near the end of 
the machine, the film in the developer end 
will continue to overdrive and thus 
produce tangled film in the developing 
section. 

Mr. Krause: It will not allow the film 
to "tangle," since the overdrive will 
admit only enough film to break the 
driving traction between the driving rollers 
and the film itself. From that point on, 
the film remains in a static position until 
normal operation is resumed. Unless 
the film was deliberately held manually 
in the drybox, this condition could not 
occur. 

Mr. Lovell: Also, do you have any filter 
in the chemical developer recirculation 
system? 

Mr. Krause: Yes, we use the Fulflo 
stainless-steel filters between our tanks 
and our recirculating pump to filter out 
the small particles. 



Tanney and Krause: Developing Machine 



267 



The Stereoscopic Art A Reprint 



By JOHN A. NORLING 



An accelerated interest in stereoscopic photography has been inspired in 
recent years by the appearance of compact, well-designed cameras and view- 
ing devices marketed at moderate prices. These new products and their in- 
creasingly widespread use support the belief that any art such as stereoscopic 
representation depends for popularity, indeed for survival, on the equipment 
available to enable people to practice the art. 



JL HAT AN APPRECIATION and under- 
standing of stereoscopic vision has ex- 
isted for a long time is evident from 
Euclid's definition: 

To see in relief is to receive by means of each 
eye the simultaneous impression of two dis- 
similar images of the same object. 

On this definition by Euclid (280 B.C.) 
the entire stereo art is based. And that 
stereoscopic representation has long been 
practiced is borne out by the stereoscopic 
drawings of Giovanni Battista della 
Porta around the year 1 600. How della 
Porta's drawings and similar drawings of 
other artists of that era were viewed 
is not known. But it is well known 
that the popularity of stereoscopic 
pictures reached a peak about 50 
years ago when the stereoscope was vir- 
tually a parlor fixture. Now, although 
the parlor stereoscope is instead a mu- 
seum piece, the art of stereoscopy appears 



Reprinted from PSA Journal, The Journal 
of The Photographic Society of America, 
Inc., Nov. and Dec. 1951, Jan. and Feb. 
1952. The author is president of Loucks 
& Norling Studios, 245 W. 45th St., New 
York 19, N.Y. 



to be flourishing again as it did half a 
century ago with one principal differ- 
ence: it is a participating art instead of 
simply a spectator art. More and more 
people are not only viewing three- 
dimensional pictures, but they are taking 
them with their own cameras. The 
popularity of stereoscopic photography 
looks as if it is here to stay. With a wider 
appreciation of its possibilities, and with 
an increasing number of camera enthusi- 
asts trying their skills at stereoscopy, its 
possible eventual widespread acceptance 
as the ultimate form of photographic 
realism may one day come about. 

Early Viewing Devices 

One of the first known viewing devices 
was the Wheatstone Stereoscope (Fig. 1), 
which was described before the Royal 
Society in 1838 by Sir Charles Wheat- 
stone sixteen years after Niepce produced 
the first permanent photographic image. 
This instrument supported a pair of pic- 
tures on easels clamped to a bar, one at 
each end. Midway between the pic- 
tures were mirrors facing them. The 
observer saw the reflected images and 
was able to fuse them provided they were 



268 



March 1953 Journal of the SMPTE Vol. 60 



-EASELS 




BAR 



Fig. 1. Diagram of the Wheatstone Stereoscope. 




Fig. 2. Illustrating the principles of 
the Brewster Stereoscope. This ar- 
rangement for prisms was used for 
pictures whose centers were farther 
apart than the normal human inter- 
ocular. 



properly spaced and aligned. But its 
unwieldiness contributed to its lack of 
popularity. 

The first practical stereoscope was in- 
vented by a Scotsman, Sir David Brew- 
ster, who introduced it in 1849 (Fig. 2). 
This was modified by Oliver Wendell 
Holmes some years later in America. 
In its early form, the Brewster stereoscope 
contained a pair of prisms which, in 
1856, were replaced by segments of a 
convex lens. The Brewster and Holmes 
stereoscopes were used for viewing paper 
stereograms. Later, several Europeans 
brought out a large variety of viewers for 
transparencies, frequently in elaborate 
cabinet form (Fig. 3). 

Just about the time Brewster's first 
stereoscope was introduced, there ap- 
peared some stereoscopic Daguerreotype 



portraits and novel lens-holding viewers. 
The accompanying illustration of one of 
these stereogram products is provided 
through the courtesy of Mr. Douglas 
Stapleton (Fig. 4). It is mounted in a 
handsome case, provided with a viewer 
containing a pair of convex lenses, with 
centers spaced 2\ in. apart. The pic- 
tures, peculiarly enough, are spaced only 
2 in. apart, introducing a little difficult} 
in viewing, but fusion can be attained. 
The name of this device is "Macher's 
Improved Stereoscope," patented March 
8, 1853. It was made in Philadelphia. 

Unfortunately, many stereoscopes had 
"eye-strain" as a built-in feature. Many 
stereograms were poor in definition and 
contrast, and sloppily mounted. These 
faults, working together, probably pro- 



J. A. Norling: The Stereoscopic Art A Reprint 



269 




Fig. 3. A variety 
of early cabinet 
stereoscopes. 

Courtesy of Robert 
N. Dennis. 



Fig. 4. A stereoscopic daguerreotype 
with lens stereoscope. 




Fig. 5. Three 
stereoscopic still 
cameras. Left, 
Richard. Center, 
Stereo Graphic. 
Right, lea. 



270 



March 1953 Journal of the SMPTE Vol. 60 



i vided one reason for the loss of popularity 
of the stereoscope. 

Amateurs who still kept alive their in- 
terest in stereoscopy had to perform a 
number of operations not required in 
ordinary photography, such as the 
transposition and mounting of the pair of 
pictures. These requirements made 
many give up the pursuit of the art. 

Oliver Wendell Holmes called a photo- 
graph "a mirror with a memory," and it 
was this pioneer in the art who christened 
the stereoscopic picture "sun sculpture." 
The "sculpture" in a stereogram is te- 
naciously remembered and details are re- 
tained that are soon forgotten in a "flat" 
picture. 

Binocular Vision 

The sensations received by the eyes are 
transmitted to the brain where, in a 
psychic assembly room, the two disparate 
images are fused together, so that we be- 
come conscious of relief in the view. But 
fusion cannot be accomplished without 
strain unless the two images are of ex- 
actly the same size and the "picture" 
axes aim at the same points. Inability 
to fuse exists in an eye disorder called 
"diplopia" in which double images are 
seen, and in "anesikonia," the term ap- 
plied when the eyes form images of differ- 
ent sizes or shapes. 

The Stereo Still Camera 

A stereoscopic camera not having 
matched lenses can be said to have 
"anesikonia," as can projectors with un- 
matched lenses. If the images on the 
screen do not align properly, the projec- 
tor has "diplopia," and the observer's 
eyes will have to perform acrobatic feats 
to attain fusion, and in the process, will 
suffer strain. Veracity in a stereogram 
exists only if it is taken properly. 

Stereoscopic cameras were made in 
scores of styles throughout the past hun- 
dred years (Fig. 5). Among the most 
widely used American stereoscopic cam- 
eras was the Stereo Graphic with a focal 



plane shutter. It took the stereoscopic 
pair on a 5 X 7 in. plate. 

The Europeans, principally the French 
and the Germans, have been very active 
in stereoscopy and have produced a 
wide range of cameras and other stereo- 
scopic devices. A well-designed French 
camera was the Richard Verascope, 
which accommodated both plates and 
roll films. Separate magazines were 
furnished for each. The one shown as a 
part of Fig. 5 took the pictures on 7 X 13 
cm plates; Verascopes were also made 
in 6 X 13 cm and 45 X 107 mm sizes. 
A single guillotine-type shutter operates 
between the elements of their two lenses. 
Shutter speeds up to y^ sec were 
provided. 

Other similar European cameras made 
pictures a good deal smaller than the 
original Richard, but some made pic- 
tures as large as 3^ in. horizontal and 
4 in. vertical. The German lea con- 
tained a valuable feature usually over- 
looked by makers of stereoscopic cam- 
eras. This was a provision for changing 
the lens spacing for principal objects at 
different distances. The lea's lenses 
move toward each other, reducing the 
lens interaxial when focusing on a near 
object, and spread farther apart when 
focusing on a far object. This is a fea- 
ture of great importance and is discussed 
later in some detail. Probably, other 
stereo camera makers will some day 
make devices to provide the highly useful 
and often necessary varying interaxial. 
If it were built into the camera, there 
would be no necessity for using smaller 
slide masks that cut down the sides of the 
image to attain satisfactory viewing, as 
we find recommended practice for close- 
ups made by many modern cameras. 

Undoubtedly, the camera most re- 
sponsible for the resurrection of interest 
in the three-dimensional art is the Stereo 
Realist. This camera was the first in the 
modern development of the stereo art 
which has features that appeal to ama- 
teur photographers, particularly those 
who have been making color transpar- 



J. A. Norling: The Stereoscopic Art A Reprint 



271 




Fig. 6. Single lens camera mounted on a slide board. 



encies on 35mm film. Like most pop- 
ular '"single-eye" miniature cameras, 
the Realist has coupled range-finder 
focusing, easy-to-learn controls, and in 
contrast with earlier cumbersome cam- 
eras, it fits the hand. 

"Slide Board" for Stereo Still Lifes 

A single lens camera can be used to 
make stereoscopic pictures of still life 
subjects by mounting one camera on a 
"slide board" (Figs. 6 and 7). Because 
only a single lens is used, both images 
will be of exactly the same size, and there 
will be none of the faults contributed by a 
pair of unmatched lenses. The slide 
board may be a simple one made of 
wood, but the precision provided by 
metal construction is much to be pre- 
ferred. Slide boards can be made as 
long as desired, but there is little use for 
one that permits a camera shift of more 
than 3 or 4 in., except for very special 
purposes. 

With a slide board, closeup stereo- 
grams can be made of objects so small and 



so enlarged, that the interaxial spacing 
may be only a few millimeters (Fig. 8). 
The slide board enormously extends the 
range of effects offered by stereoscopic 
photography, and it is an economical and 
convenient method for the person who 
would like to start in stereoscopy without 
first buying a two-lens camera. 

Kennedy Stereo Camera 

Even without a slide board, the single 
lens, if its aperture is large enough, can 
be employed to produce stereograms of 
excellent quality. Its usefulness seems 
to fall only in the still-life field, since the 
effective aperture employed for each 
image is of a rather low order. The 
system is simplicity itself, and was de- 
signed and built by Prof. Clarence 
Kennedy of Smith College, in coopera- 
tion with Dr. Edwin H. Land and Otto 
E. Wolff (Fig. 9). Professor Kennedy 
had employed it principally to obtain 
large stereograms of sculptures. Its 
diagrammatic representation shows that 
only a part of each lens is used for form- 



272 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 7. A simple slide board: A, sliding 
member; B, camera screw (^-20); C, 
main body; D, bushing for tripod screw; 
E, spirit level; F, holes to clear screw 
driver so camera may be attached to 
A from below; G, angle brass; H, brass 
strip; J, stops; K, pointer; L, scale (paper 
or cloth tape). 




Fig. 8. Stereogram of a watch movement (lens interaxial 5 mm). 




LEFT DIAPHRAGM 




RIGHT DIAPHRAGM 



Fig. 9. Illustrating the principle of the 
Kennedy stereoscopic single lens camera. 



J. A. Norling: The Stereoscopic ArtA Reprint 



273 



ing each member of the stereoscopic pair. 
If the lens has a large enough diameter to 
accommodate the desired result, the two 
individual lens "stops" can be separated 
to the required spacing called for by the 
geometry of the stereogram's planned 
use. A lens of large diameter, 75 mm or 
more, must be used. Its focal length is 
determined by the type of stereograms 
to be made. A large diameter demands 
a lens of long focal length. Conse- 
quently, this particular process is limited 
to still stereograms. Because of the 
large size of many stereograms used this 
way, projection would introduce many 
problems. Viewing boxes are used in- 
stead. 

Stereographic Drawings and Paintings 

A more or less obscure form of the 3-D 
art is the rendering of Stereographic 
drawings and paintings. As a teaching 
aid, Prof. John T. Rule of the Massa- 
chusetts Institute of Technology has 
introduced such novel drawings for 
celestial navigation and solid geometry 
(Fig. 10). They provide the student an 
unusually effective portrayal of spatial 
relationships. Some of Prof. Rule's 
drawings have been published in vecto- 
graph form as an adjunct to classroom 
texts. The vectograph method is dis- 
cussed later in this article. 

In the illustration field, Paul H. Stone 
has produced a number of striking 
stereograms. With his kind permission, 
we are able to publish one of his remark- 
able creations which he has entitled 
"Clinical Study of How Stereo Registers 
to the Brain" (Fig. 11). Mr. Stone has 
a theory which makes interesting reading 
on the subject of how the eyes work : 

"Our two eyes scan every pin-point of 
a scene, as rapidly as a television beam ; 
from top to bottom, side to side, and, 
most important, from near to far. It is 
less the focus of each cornea than the 
convergence of the two beams of vision 
which telegraphs a sense of depth, of 
distance to the brain. 

"I believe the actual muscular pull on 



the eyeballs, whether they are crossedj 
toward each other or straightened toi 
parallel, and infinity, is the recording de- 
vice which informs the brain about thei 
length of the convergent triangle. Hence 
the brain can infer with reasonable 
accuracy, depths, from a fe\v inches to 
the better part of a mile." 

Versatile Stereo Camera 

While the amateur has available a 
\\ide selection of stereoscopic cameras 
and accessories, there are no well- 
designed cameras for the professional. 
If he wants to go into stereoscopy, viri 
tually the only thing he can find is an old 
Stereo Graphic in a second hand store, 
or if he is fortunate, one of the excellent 
leas. But neither of these has all the 
features required by the serious worker. 

A professional stereoscopic camera 
should contain, as principal features, a 
means for varying the lens interaxial and 
a means for converging the picture axes 
on any desired plane. 

Probably the easiest way to show what 
a versatile stereoscopic camera is like is to 
describe one of several types designed by 
the author (Figs. 12, 13a and 13b). 
The diagram shows the plan view layout. 
The assembly comprises two camera 
bodies mounted on a common base. 
Each has its own lens, but the lenses are 
mounted in a unique way, pointing off at 
right angles, one to the left, the other to 
the right. In front of each lens is a first 
surface mirror whose face is 45 to the 
lens axis. A righthand-lefthand screw 
provides a means for moving the camera 
bodies together or apart, thus controlling 
the interaxial spacing. This spacing 
can be varied from 1^ to 6 in., permit- 
ting satisfactory setting for objects as 
close as 4^ ft, using 7-in. lenses. The 
assemblies which carry the lenses and 
front mirrors are moveable for a small 
distance forward and backward. This 
makes it possible to converge the picture 
axes to any plane desired. The images 
are framed within metal masks which 
are mounted so they can retract for the 



274 



March 1953 Journal of the SMPTE Vol. 60 





JOHN T. RULt 



Fig. 10. A stereographic drawing illustrating a geometrical 
problem. Courtesy Prof. John T. Rule. 




Fig. 11. Stereoscopic painting. Courtesy Paul H. Stone. 



MOVEABLE FOR . 

CONVERGENCE t 

CONTROL \ 



. MOVEABLE FOR 
t CONVERGENCE 
J CONTROL 

INSIDE 
IRROR 



J FRONT MIRRORS LJ 
LE. LENS R.E. LENS 




FILM 



Fig. 12. Diagram illustrating the arrangement of components of a 
special still picture stereoscopic camera. 



J. A. Norling: The Stereoscopic Art A Reprint 



275 




CONTROL 
Fig. 13a. A versatile still stereoscopic camera, one of several designed by the author. 



insertion of the film holders dark slide. 
These masks accurately define the pic- 
ture areas, and if the convergence con- 
trol is properly used, the masked margins 
provide precise registration for mounting. 
It should be emphasized that precise 
alignment of the pictures is an absolute 
requirement for projected stereograms. 
If not precisely aligned, the 3-D pictures 
cannot be viewed with complete visual 
comfort, and the goal of the serious 
workers is and should be nothing short of 
complete visual comfort in viewing. 

The mask on the right has a small hole 
close to the picture area that permits a 
cue mark for the purpose of quickly 
identifying it, thus facilitating the 
assembly of the pairs. This camera is 
used mostly for making color 3-D slides, 
although it has been used to make master 
black-and-white negatives for vecto- 
graphs and anaglyph stereograms. 

One arrangement that can be em- 
ployed to attain the variable interaxiai 
feature is illustrated in Fig. 14. In this 
arrangement a partially reflective, par- 
tially transmittive mirror is used, one 
camera being placed behind the mirror, 
the other being placed so that it receives 
the image by reflection from the par- 
tially silvered first surface of the mirror. 



This method permits the interaxiai to 
be varied from zero to the limit per- 
mitted by the size of the mirror. Of 
course, the "left eye" camera (in the 
arrangement shown) will have its image 
reversed left for right, but that is unim- 
portant because the negative can be 
turned back to normal in printing. 

The mirrors should be the pellicle 
type, such as are used in some color- 
separation still cameras. Satisfactory 
results have been obtained with thin 
glass mirrors, -% in. thick or there- 
abouts. A thick glass mirror cannot be 
used because a second reflection will 
show in the image picked up by the 
mirror. If a thin glass mirror is used, 
the secondary reflection will not be 
separated far enough from the primary 
image to become discernible if the mirror 
is mounted at about 57, as shown in the 
diagram. 

If a pellicle mirror is used it must be 
mounted in a frame that will afford the 
facility for stretching it taut. 

Taking Stereo Still Pictures 

The stereo still photographer is con- 
fronted with just about the same prob- 
lems as the conventional still photog- 
rapher, plus a few extra problems that 



276 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 13b. Rear view of camera showing the picture mask. 



MOVEABLE,- PARALLEL l 
TO PLANE OF 

MIRROR | 

k-H 




Fig. 14. Illustrating an arrange- 
ment that provides a variable lens 
interaxial using a partially trans- 
mitting pellicle mirror. 




PELLICLE 
MIRROR 

50% REFLECTING - 
50% TRANSMITTING 



RE CAMERA 



are peculiar to stereo. Interocular dis- 
tances, interaxial distances and converg- 
ence problems are some of the extras 
which lie only in the province of stereo. 
But they are soon mastered and the 
pleasure derived from their mastery is 
rewarding indeed. 

A difference in focal length between 
the lenses of a stereo camera of more than 
0.5% will produce images differing 



enough in size to cause eyestrain when 
pictures are projected. 

If glass filters arc used on the lenses, 
there may be enough "wedging" in the 
filter assembly to throw the picture axes 
out of line. This is not serious if the mis- 
alignment is along the horizontal, but it 
may be significant if it is along the verti- 
cal, and when the pictures are projected, 
it may be found that they have disturbing 



J. A. Norling: The Stereoscopic Art A Reprint 



277 



LEFT EYE 


RIGHT EYE 


* f 


~~! ' ' 


(tj) 

i" 


1 J O III 

J 


Q 


| / / \(^) i' 


\ ' v 


! 1 f\ \\ "/ 


\ x 
\ \! 

\ .\ 


/ / \ i // 

/ /FOREGROUND, . ' \ / 1 
/ PLANE \ \ 1 / / 

f. ' \ \ / / 


\ IV 


! / \\ V 1 


\ ! X 


\ ! \ ' \/\ ./ 


\ 1 \ 


1 / ^1 '^ \ j 


V 


V' V V 


1 


1 / \ 


1 


i ' i 


Fig. 15. Left, the 


side margin effect with nonconvergent lenses. 


Right, the effect i 


,vith lenses converging on a foreground plane. 



vertical displacement between them. 
This "vertical" may not produce a vivid 
consciousness of eyestrain, but it may 
very well produce enough effect to tire 
the observer after a time. As J. M. 
Dalzell has said, "Any sensation of visual 
lassitude, discomfort or even slight pain, 
is a psychological and mental affair. It 
is the penalty for exceeding certain geo- 
metric limitations which Nature has im- 
posed upon the use of two eyes simul- 
taneously. Its cause is cumulative and 
lies in the breaking of one or more of a 
few exceedingly simple rules." 

To avoid eyestrain, all elements in 
every plane must have crisp definition 
such as good eyes provide in their wan- 
derings from plane to plane when looking 
at a scene. Out-of-focus or "soft focus" 
areas, whether in foreground or back- 
ground, are an abomination that pre- 
vents visual comfort in viewing the 
stereogram. 

Since the normal 2^-in. lens inter- 
ocular ordinarily should be used on 



objects no closer than about 10 ft. ir is 
apparent that a properly made stereo- 
gram can be taken only in a camera 
having a variable interaxial. Of course, 
some shots may benefit from a larger 
interaxial than normal, particularly in 
cases where an increased depth is con- 
sidered a dramatic embellishment. By 
increasing the interaxial to several times 
normal, the depth of such stereograms 
can be startlingly exaggerated. 

The variable convergence feature is a 
"must" on any versatile stereoscopic 
camera because it permits the arrange- 
ment of each picture within the desired 
area. To be able to converge to any 
plane, preferably in front of the nearest 
object, eliminates the marginal disturb- 
ances that occur when the lens axes are 
parallel (Figs. 15 and 16). Such mar- 
ginal disturbances are particularly an- 
noying when they exist in the projected 
stereogram. The ideal stereogram 
should be a "real-life" view as seen 
through a window. Therefore, no ob- 



278 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 16. Illustrating how nearby object images are cut off when 
image axes do not converge on objects. 



ject should be permitted to invade the 
plane where the window exists if that 
object "touches" any part of the window 
frame. If it does, the result is a dis- 
tracting border which gives non-coinci- 
dental image reconstitution. 

If the "window" is at infinity, as it will 
be with parallel lens axes, the illusion 
will be that foreground middle distance 
and part of the far distance planes are 
between the window and the observer, a 
totally abnormal effect. 

To create a proper window illusion the 
masks framing the images must be 
identically alike in shape and size and 
their horizontal margins must be level. 

Masks can be of any shape. For in- 
stance a keyhole mask will provide a 
stereoscopic view through a keyhole, an 
anomalous effect since keyholes are so 
small that real-life peering through one is 
limited to one eye. 

Where objects do not touch the win- 
dow frame (the picture margins), they 
can be permitted to come through the 
window and even made to seem sus- 
pended in space between the screen and 
the observer. In table-top photography, 
one can produce this effect by mounting 
the subject on glass, being sure it is well 
inside the picture area and not touching 



its margins at any point. Convergence 
is made to occur behind the subject. 
Viewing comfort is then readily achieved. 

Convergence is best accomplished by 
moving the lenses, not by "toeing-in" the 
camera bodies. Toe-in introduces a 
"keystone" distortion that makes the 
right edge of the left-eye picture shorter 
than the left edge, and the left edge of 
the right-eye picture shorter than its 
right edge (Fig. 17). This kind of dis- 
tortion may not be considered important, 
and will be tolerated by some people, 
but it is one of the things to be avoided if 
the stereo worker wants to insure that his 
projected stereograms provide complete 
visual comfort. 

To make a satisfactory stereoscopic 
picture, the camera must be level and to 
project the images successfully, there 
must be no vertical displacement be- 
tween them and they should not "ro- 
tate" one with the other. In other 
words, the angular alignment has to be 
the same for both members of the pair. 

We may work our interaxial problem 
out by "rule of thumb." Stereograms 
that are perfectly acceptable, if properly 
framed in projection, can be made by 
taking pictures only of objects further 
than 50 times the interaxial employed. 



J. A. Norling: The Stereoscopic Art A Reprint 



279 



DISTORTION 
IN LEFT EYE IMAGE 




DISTORTION 
IN RIGHT EYE IMAGE 




Fig. 17. Left, the distortion resulting from the "toe-in" of camera lenses 
Right, a method of convergence to overcome distortion. 



(This is for camera lenses of the medium 
angle usually employed.) This rule 
limits the nearest object to a distance of 
10 ft with an interaxial of 2^ in. It is 
violated to an extreme if 2^- in. is used 
for a 5-ft distant object, where the indi- 
cated interaxial is 1^ in. 

Modern Viewing and Projection 

Bringing out a modern camera for 3-D 
use was not enough to stimulate the re- 
birth of interest in the stereoscopic art. 
There had to be some device enabling 
people to see the 3-D picture. The 
manufacturer of the Stereo-Realist ap- 
preciated this from the outset, and ac- 
cordingly brought out an individual 
viewer. Similar camera-viewer com- 
binations have appeared since the Real- 
ist made its debut. But an individual 
viewing device is hampered by an 
obvious shortcoming; it limits the 
pleasure of looking at a 3-D picture to 
one person at a time. This short- 
coming does not apply, however, to 
projected stereo pictures. Shortly after 
the introduction of camera-viewer com- 
binations, a 3-D projector became 
available and it enables many people at 
the same time to see and enjoy the same 



stereo pictures. There are now several 
stereoscopic slide projectors on the mar- 
ket, all more or less alike. 

In addition to the added enjoyment it 
provides, the projection of stereograms 
to fill a large screen does one more thing 
that the individual and personal stereo- 
scope never can accomplish, and that is 
to create the feeling that the scene is big. 
While it is perfectly true that the picture 
on the screen may be of the same angle of 
view as the same picture mounted in a 
stereoscope, and that theoretically there 
should be no physical reason why the. 
scene elements should not appear the 
same in each, looking through a stereo- 
scope very often gives one the feeling 
that he is looking at a miniature, and he 
has that feeling whether or not the scene 
was photographed normally. Perhaps 
the fact that the picture is itself small in 
dimension gives rise to the feeling that 
the original scene must likewise be small. 

Most stereograms made today are in 
the form of slides which are looked at 
through a hand-held viewer or seen as 
projected images. The hand-held, self- 
illuminated viewer is a familiar device. 
Some viewers, including the Realist, 
have provision for a variable lens spac- 



280 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 18. Illustrating the principle of polarized light stereoscopic projection. 




Fig. 19. Various types of Polaroid 3-D viewers. 



ing a valuable feature that con- 
tributes to visual comfort. The Stereo- 
Realist image is |- in. wide and if in. 
high, but there are other mask sizes such 
as 1 in. wide by -J in. high, and 1-|- in. 
wide by J~| i n . high. All can be 
mounted in the standard If X 4 in. 
glass, which will go into the slide carriers 
of the new stereoscopic projectors. For 



professional type slides, each member of 
the stereo pair is made up in standard 
3^ X 4 in. glass, and put into a holder 
which accommodates the two members 
side by side. 

Projection of stereo slides requires the 
use of polarizing filters, one for each lens, 
an aluminum-surfaced screen, and polar- 
izing 3-D viewers (Figs. 18 and 19). 



J. A. Norling: The Stereoscopic Art A Reprint 



281 




The projection filters are arranged 
with their polarization axes slanting 45 
from the vertical, one slanting to the 
right, the other to the left. The viewers 
have polarizing filters which also "slant" 
in the same manner. Consider the light 
from the righthand lens, with a polariza- 
tion axis slanting toward the left. The 
light reflected from the screen can be 
seen through only the right-eye polarizing 
filter, because it too has the same left- 
hand slant while the left-eye filter has a 
righthand slant, 90 from that of its 
mate. Thus the "right-eye" light is 
blocked from reaching the left eye, and 
the "left-eye" light is blocked from 
reaching the right eye. 

The reason for using an aluminum- 
surfaced screen is that this type of surface 
will not disturb the polarization of the 
light. A fabric or glass-beaded screen 
will cause the polarization to disappear, 



Fig. 20. Stereoscopic 
views made by high- 
speed flash, 1 jusec ex- 
posure at f/9 in Ekta- 
chrome. 



the light to become ordinary light again. 
The result is that the two images will be 
seen by both eyes, and will look jumbled 
like a double exposure. 

Plastic screens are now available for 
rear projection of polarized light stereo- 
grams. Rear projection adds a brilliant 
quality quite different from the reflection 
method, although a "hot spot" may be 
disturbingly apparent unless arrange- 
ments are made to keep the spectator's 
eyes and the projection axis from lining 
up. 

For extremely dramatic stereogram 
presentations, a picture 18 in. wide can 
be thrown by a projector containing two 
1200-w bulbs and a pair of 18-in. //3.8i 
lenses. With a picture this wide, an 
audience is usually given to "oh-ing" 
and "ah-ing" with delight and surprise 
as each successive stereogram appears on 
the screen. 



282 



March 1953 Journal of the SMPTE Vol. 60 



Hyperstereoscopy 

Hyperstereoscopy is the term applied 
when an interaxial base several times 
normal is used. It has often been em- 
ployed in mountain photography and 
serves admirably to reveal distant de- 
tails in relief. The base employed is 
often 100 yd or more. Care must be 
taken not to include any foreground, 
otherwise one image will contain ele- 
ments not present in the other, and it will 
be impossible properly to fuse the stereo- 
gram. Since mountains can usually be 
depended upon to stand still for a long 
time, there is time to set up the camera 
and to transport it from one extreme of 
the interaxial base to the other. But 
clouds are not so considerate in "staying 
put" as mountains, so the exposures 
should be made on a cloudless day. 

Hyperstereograms, properly made, re- 
quire that certain rules for determining 
the interaxial be followed. The sim- 
plest one is 



d X D 
D-d 



-T- 50 



I where d is the distance to the nearest 
object, D the distance to the farthest and 
50 the divisor whose use assures success. 

Keeping within the limits prescribed 
will give the stereograms the most vivid 
relief possible without eyestrain in view- 
ing. 

Hyperstereoscopy has been employed 
to make stereograms of bodies in the 
solar system where even the mean diam- 
eter of the earth often proves insufficient 
as an interaxial base. Stereograms of 
most planets have been made and in- 
teresting data has thus been obtained. 
Included has been the discovery of a 
planetoid, which was found as the result 
of this type of stereoscopic survey, and it 
was appropriately named Stereoscopia 
in tribute to the method. Stereograms 
of the moon have been made which re- 
veal intimate details of its craters and 
seas. Some of the most remarkable have 
been those of Saturn and its rings, one 



pair of which required the movement of 
Earth through space that was the equiva- 
lent of an interaxial base of about 1 ,000,- 
000 miles, which is the distance traversed 
by our planet in its orbit during a 24-hr 
period. Other stereograms of this planet 
made with an even greater base disclosed 
clearly the separation of its unusual rings 
one from another and from the planet. 

A word of warning concerning hyper- 
stereoscopy : it does not seem to produce 
satisfactory results for close-up objects, 
and certainly will not do so if such stereo- 
grams are projected. This is so for close 
objects because the angle formed be- 
tween the two lenses at the base and the 
object is so great that it throws distortion 
into the images. For instance, photo- 
graphing a golf ball at a distance of one 
foot and using a wide interaxial will pro- 
duce a stereogram which makes the golf 
ball appear egg shaped, and golf balls 
having this shape give neither the player 
nor the viewer any pleasure (Fig. 20). 

High-speed stereoscopic photography 
is often employed for special purposes. 
Such stereograms often reveal things not 
readily apparent in a flat picture. They 
are particularly useful in the study of 
machine elements in motion and for 
other kinds of research. A series of 
three high-speed "strobe" shots were 
made for United States Rubber Com- 
pany to prove the behavior of a golf ball. 
The illustrations show the impact of the 
club head, the "flattening-out" of the 
ball before leaving the tee, and the ball 
in its flight a few inches ahead of the club. 
These shots were made on Ektachrome 
film at a speed of one-millionth of a sec- 
ond, using one of Prof. Harold E. 
Edgerton's newer flash units. Prof. 
Edgerton and Henry Lester worked to- 
gether in obtaining these shots. The 
projected pictures showed no "egg- 
shape" distortion which would have re- 
sulted if an interaxial larger than called 
for by the object distance had been em- 
ployed. 

No graphic means, besides the stereo- 
gram, can substitute for the re-creation 



J. A. Norling: The Stereoscopic Art A Reprint 



283 




Fig. 21. Interlocked 16mm cameras for 

stereoscopic photography. Courtesy Carl 

Breer. 

of the "real" in a still-life, and in stereo 
movies realism reaches the ultimate, for 
they can include movement, color, and 
action as well as depth. 

The principles employed in photo- 
graphing and projecting stereoscopic 
slides also apply to stereoscopic motion 
pictures. The same fundamental re- 
quirement that each eye sees only the 
picture intended for it also applies to the 
moving stereogram. 

Stereoscopic Motion Pictures for the 
Amateur 

The experimentally inclined camera 
enthusiast will get a lot of thrills in mak- 
ing and projecting stereoscopic movies. 
Some have done it, others are planning 
to enter this new field. To encourage 
those who are thinking about it, I shall 
cite the experiences of one avid fan, 
Carl Breer, Vice-President and Director 
of Research of Chrysler Corp. He has 
kindly provided the accompanying 
photographs illustrating his camera and 
projector arrangements. 

His first camera hook-up consisted of 



two 16mm Magazine Cine-Kodaks, 
mounted so their lens axes could con- 
verge on any desired plane (Fig. 21). 
He selected Magazine Cine-Kodaks be- 
cause they were the most adaptable at 
that time (1940) for mounting at the 
normal 2^-in. interaxial. Since this 
camera arrangement requires two strips 
of film, he found it necessary to rig up 
interlocked projectors to show his pic- 
tures. Accordingly, he selected two 
Bell & Howell 16mm projectors and in- 
terlocked them mechanically (Fig. 22). 

Carl Breer has had a lot of fun with his 
stereoscopic motion pictures, not only in 
taking pictures and exhibiting them to 
his family and friends, but also in devising 
the ways and means. 

35mm Stereoscopic Motion Pictures 

It seems incredible to many of us who 
have worked with 3-D pictures that the 
vast motion picture industry does not 
have an extensive stereoscopic research, 
engineering and development program. 
Occasionally, at the meetings of the 
Society of Motion Picture and Tele- 
vision Engineers, the subject of 3-D 
motion pictures is introduced. There is 
usually a remarkable response from the 
members present, and also from the press. 

The art of stereoscopy has "sex appeal" 
but it seems to have excaped the con- 
centrated attention of most of the people 
in the Hollywood area. The men in the 
drivers' seats of the movie industry have, 
for the most part, failed to have a vital 
personal interest in and understanding of 
3-D movies. That the industry could 
use something to combat television's 
capture of more and more of the theater 
audience is undeniable. Stereo movies 
might well induce people to return to 
their former favorite amusement. But 
the return is likely to come about in the 
mass only if the film theater gives them 
something they can't get on a 17-in. 
television tube, namely the ultimate in 
photographic realism, the stereoscopic 
movie in full color, with all the dramatic 
possibilities that are only waiting to be 



284 



March 1953 Journal of the SMPTE Vol. 60 




Fig. 22. Two 16mm projectors interlocked for stereoscopic projection. 



appreciated. The enthusiastic public 

reception given some earlier stereo movies, 
] and the dollar profits from these movies 
I are a matter of record. Newer, better 
i stereo techniques are now available, and 

the reason for introducing them was 

never more pressing. 

The Anaglyph 

One of the early and noteworthy 
theatrical exhibitions of stereoscopic 
motion pictures occurred in 1924, when 
J. F. Leventhal produced a few "shorts" 
utilizing the anaglyph process. There 
followed an 1 1 -yr lull in the use of stereo- 
scopic films. Then, in 1935, Loucks & 
Norling Studios and Mr. Leventhal 
jointly produced a series of short films 
again employing the anaglyph principle, 
this time in talking picture form. These 
films, which were called Audioscopiks, 
were released by Loews, Inc., and proved 
to be some of the most successful short 
subjects ever issued, winning not only 
domestic acceptance, but an unprece- 
dented play in the foreign field, notably 
in France, Spain and Great Britain. 



That their success should have indicated 
further pursuit of the anaglyph process 
seems logical. But the producers had, 
from the beginning, realized the in- 
herent limitations of the anaglyph proc- 
ess and concluded that films exhibited 
by that process would only be adequate 
as novelties and would never be toler- 
ated for full-length feature releases. 
This conclusion was arrived at by a rec- 
ognition of the visual "insult" resulting 
from the projection of one color to one 
eye and its complementary to the other. 
This sort of delivery of images, one color 
to one eye, another to its mate, produces 
"retinal rivalry" and brings on physi- 
ological disturbances that may induce 
nausea in some observers if they look at 
the anaglyph movie longer than a few 
minutes. 

Since this process the anaglyph 
has played an important role in the ad- 
vance of the stereoscopic art, it would be 
well to describe it here briefly. Its in- 
vention is credited to Ducos du Hauron, 
who applied it in 1895, although there 



J. A. Norling: The Stereoscopic Art A Reprint 




Fig. 23. Above, Bell & 
Howell 35mm camera inter- 
locked for stereoscopic photog- 
raphy. Simultaneous focus- 
ing of the lenses is effected by 
rack and gears. Below, Bell 
& Howell 35mm interlocked 
cameras provided with three- 
color filter wheels for making 
stereoscopic color-separation 
negatives by the successive- 
exposure method. Used only 
for "stop-motion" work. 



286 



March 1953 Journal of the SMPTE Vol. 60 



is some evidence that its possibilities had 
been explored many years before that. 

In one form, the anaglyph images are 
on two separate films. One member of 
the stereoscopic pair is projected through 
a filter of one color, the other through a 
filter having a color complementary to 
that of the first. In another form, the 
one that was used for Audioscopiks, the 
anaglyph images are printed in comple- 
mentary colors directly on film and pro- 
jected in a standard projector without 
filters. 

The projected images are viewed with 
spectacles having windows of the same 
colors as the colors on the screen. Red- 
orange for the right-eye filter and blue- 
green for the left are often used. The 
right-eye red-orange filter in the viewing 
spectacle renders the blue-green right- 
eye image in monochrome and the left- 
eye blue-green filter renders the red- 
orange left-eye image also in mono- 
chrome. Since dyes and pigments 
hardly ever are capable of transmitting 
only the color they are supposed to trans- 
mit, there is rarely a complete "cutting" 
of one color; some of it always comes 
through so that part of the blue-green 
image which is supposed to be blocked 
by the blue-green spectacle filter leaks 
through, producing a "ghost" image. 
So, in reality, one eye sees a part of the 
image intended for the other, the "part," 
of course, being defined as a very dim, 
but still discernible remnant of the whole 
"other-eye" image. 

Good picture quality has never charac- 
terized the anaglyph. This and other 
shortcomings make it eligible for discard 
as a practical system for motion picture 
features. 

Since the introduction of Polaroid 
light-polarizing filters, it is possible and 
practical to substitute these for the red 
and green filters of the original anaglyph 
process. Strictly speaking, the polarized 
light method may be defined as another 
form of the anaglyph. Actually, Pola- 
roid Stereoscopy would be a good name 
for it, since Dr. Edwin H. Land, head of 



Polaroid Corp., invented the first prac- 
tical and efficient synthetic polarizer 
which hastened the increasingly wide- 
spread use of the present satisfactory 
methods of stereoscopic projection. 

Stereoscopic Motion Pictures at the 
New York World's Fair 

The first large-scale public exhibition 
of a 35mm stereoscopic motion picture 
with excellent picture quality took place 
in 1939 at the New York World's Fair. 
That year a black-and-white film was 
shown. The following year a similar 
subject was exhibited in Technicolor. 
More than five million people saw these 
films,* and they're still talking about 
them. Some of the production and exhi- 
bition problems posed by these pictures 
are interesting to consider. 

The camera assembly for the black- 
and-white picture consisted of two Bell & 
Ho well professional 35mm cameras 
mounted so that one was "upside down" 
in relation to the other (Fig. 23). This 
was done so that the lenses could be 
brought close together. Even with this 
arrangement, the interaxial was not 
ideal. It was fixed at 3| in. although 
calculations showed that some scenes 
actually required as close as 1^-in. 
interaxials. But no such camera was 
available then, nor was there time to 
have one built. However, a complete 
set of matched lenses of different focal 
lengths effected a quite satisfactory 
compromise with the ideal. 

The greater part of the picture was a 
sort of fantasy, showing the parts com- 
prising a Plymouth car dancing around 
and assembling themselves. Their 

movements were in synchronism with 
music and required the use of "stop 
motion" photography; that is, "one 
frame-at-a-time" snooting. But a sub- 
stantial part of the film contained "live 
action" shots taken in the foundry and 
shops and along the assembly line. The 
narrator for the film was Major Bowes of 

* Produced by the author. 



J. A. Norling: The Stereoscopic Art A Reprint 



287 




Fig. 25. Above left, the Loucks and 
Norling Studios' 35mm stereoscopic 
motion picture camera. Camera is 
in shooting position. Above right, 
front of the 35mm stereoscopic camera 
showing the variable interaxial system 
which has a range from 1^ in. to 4 in. 
Right, the 35mm stereoscopic camera 
racked over for lining up the scene 
and focusing. The binocular view- 
finder has interocular adjustment. 



Fig. 24. Mechanical 
interlock of two 
35mm projectors. 




288 



March 1953 Journal of the SMPTE Vol. 60 



Amateur Hour fame. He appeared in 
"live action" in one sequence in which 
he spoke. This was the first "live 
action - live dialogue" shot ever made 
in a stereoscopic presentation. It 
created some difficult problems since the 
cameras would not fit into any available 
studio "blimps." However, the se- 
quence was shot without any parasite 
camera noises being recorded, thanks to 
the intelligent help rendered by Walter 
Hicks, then of the New York Fox- 
Movietone Studios. 

Since the Chrysler film was shot in a 
two-camera setup, and no special photo- 
graphic and projection facilities for 
single-film handling were available, it 
was necessary to project with two pro- 
jectors (Fig. 24). A rather complex 
selsyn motor drive was used for interlock, 
although a much simpler synchronization 
could have been attained by a straight- 
! forward mechanical linkage, such as was 
used for the Pennsylvania Railroad 
stereoscopic movie display at the Golden 
Gate International Exposition in San 
Francisco in 1940. 

The dual projector system used at the 
New York and San Francisco Fairs is 
substantially the same as that recently 
on exhibition at the Festival of Britain. 
According to press reports, it is also the 
same system which has recently been 
demonstrated by the Natural Vision 
Corp. of Hollywood. 

A Technicolor film, using the stop- 
motion technique as well as live action 
shots on monopack was our next stereo 
production. For the stop-motion se- 
quences, a unique filter attachment 
was arranged in front of the camera 
lenses. The filters were mounted on 
wheels which rotated together. Color 
balance was attained by making sectors 
having angular dimensions calculated 
to pass the quantity of light required for 
each color and as demanded by the sensi- 
tivity of the film. The "A" (red) 



filter passed light to which the film was 
more sensitive than that passed by the 
"B" (green) and "C5" (blue) filters. 
Consequently, the red filter had the 
narrowest opening of all and the "C5," 
to whose transmission the film was least 
sensitive, had the widest opening. The 
exposures were made by the alternate 
frame method of color separation. 
Three frames, one the red record, one 
the green, and one the blue, were made 
instead of one frame as in ordinary 
photography. This procedure is fol- 
lowed in the photography of animated 
cartoons. 

These separation negatives were used 
by Technicolor to make the printing 
matrices from which the dye imbibition 
prints were produced. 

It has always been the author's opinion 
that the stereoscopic camera for profes- 
sional use should be built to take the 
images on two separate films. This is to 
afford the greatest flexibility in the studio 
and to permit the use of short focus 
lenses and to facilitate the making of 
optical effects in the duplicating proc- 
esses. One such camera was built 
(Fig. 25). It contains the features 
deemed essential to a versatile camera. 
The most important are a variable in- 
teraxial and a convergence control, but 
important too is a binocular finder show- 
ing in miniature a three-dimensional 
view of the scene to be photographed. 
Visual inspection during focusing seems 
superior for stereoscopic work and focus- 
ing is easier when the view is seen in three 
dimensions. The binocular viewfinder 
has an additional advantage : it enables 
the cameraman to compose the scene 
stereoscopically using the interaxial and 
convergence controls, manipulating them 
until he gets the best possible arrange- 
ment. He can increase the interaxial if 
he wants to increase the apparent depth 
of the scene. He can reduce it if nearby 
objects demand it. 



J. A. Norling: The Stereoscopic Art A Reprint 



289 



Dual Images on One Film 

Systems for stereoscopic films using 
dual images side by side or one above 
the other have also been proposed. 
One of the problems in the two-image ar- 
rangement, whether in tandem or side 
by side, is the loss of light, because the 
light-covering circle covers a large area 
around the area occupied by the two 
images. 

The ordinary circular light spot from 
the projector arc spills light all around 
the images. This condition can be im- 
proved upon by a light condensing sys- 
tem having a cylindrical lens element. 
Then the light spot becomes oval in- 
stead of round (Fig. 26). 

The Newcomer Anamorphoser 

Another method is to introduce an 
optical device on the camera to compress 
the images in one direction, and a 
similar device on the projector to ex- 
pand them back to normal proportions. 
Such an optical device is called an "ana- 
morphoser" (Fig. 27). Several types 
have been constructed, but it remained 
for Dr. H. Sidney Newcomer to design 
one that does not introduce serious aber- 
rations and have other optical handi- 
caps. The Newcomer Anamorphoser is 
capable of effecting a compression of the 
image to almost % and an expansion of 
almost 1^ times. 

The Beam Splitter 

Among the methods suggested for the 
employment of a single film to carry 
the two images is the "beam splitter" 
in one form or another (Fig. 28). The 
device has two pairs of mirrors (or 
prisms) placed in front of the lens and 
arranged so that the pair on the left 
will cause the left-eye image to be selected 
for projection to the screen and the right- 
hand pair will do the same for the right- 
eye image. It is a simple device and easy 
to use. 

The beam splitter is a device that does 



exactly what its name implies; it splits 
the light beam into two parts. Hence, 
the intensity of each part cannot be 
greater than half of the whole beam. 
But in addition to light loss, it has an- 
other drawback. The pictures overlap 
considerably, making it impossible to 
mask them to a stereoscopic window. 
The window must be artificially pro- 
duced by a black border on the screen to 
absorb spill-over light. Another short- 
coming: The camera lens works at 
something less than half the //stop setting 
shown on the lens. This means more 
than twice the amount of light required 
for conventional photography. When 
it comes to shooting interiors, this added 
light requirement proves to be an eco- 
nomic disadvantage of the beam splitter 
method. There is a corresponding light 
loss in projection, and here the loss is 
even more significant. Take the loss 
inherent in the beam splitter, add that to 
the loss in polarization and you find that 
you're getting about one-twelfth the 
light that you had when you projected 
the full frame in the conventional way. 
Another disadvantage of the ordinary 
beam splitter is the picture proportions it 
gives a narrow and tall picture, 
certainly inappropriate for stereoscopic 
representation which is so well suited for 
panoramic views. 

Another proposed device has dual 
lenses producing pictures side by side. 
There is no light loss in the camera, 
since two lenses are used and the window 
for each picture is quite sharp. How- 
ever, there is considerable loss in pro- 
jection if the attachment is used on a pro- 
jector not provided with a special con- 
denser system. If the standard propor- 
tions are retained, each image is less than 
one-fourth the area of the full frame. 

Another method which has been pro- 
posed for simultaneous projection, is the 
arrangement of images with one member 
above the other. Special projectors 
would be required. 



290 



March 1953 Journal of the SMPTE VoL 60 




WITH ORDINARY 
CONDENSERS 



WITH A CYL- 
INDRICAL 
CONDENSER 



LIGHT, 
LOSS 

Fig. 26. A light-condensing system that improves the light 
distribution to dual images. 



IMAGES EXPANDED 
VERTICALLY BY 

CONTRACTED ANAMORPHOSER 
IMAGES ON 
FILM \ 




Fig. 27. Diagram of the "Anamorphoser" principle applied 
to motion picture stereoscopy. 





Fig. 28. Diagram of one arrangement of 
the "beam splitter." 




J. A. NorUng: The Stereoscopic Art A Reprint 



291 



LEFT EYE 



RIGHT EYE 




Fig. 29. The light and dark periods in "eclipse" stereoscopic projection. 



Sequential-Frame Projection 
"Eclipse" System 

From time to time, the alternate 
projection of the members of a stereo 
pair has also been proposed. In this 
system, the right-eye image, for instance, 
is projected first, then the shutter inter- 
rupts the light beam while the film 
moves down to position the left-eye 
image. Thus there are periods of flicker 
that occur at different times for each 
eye. If we break this sequence of events 
down, we find that the first light period 
has a value of 12.5% of the complete 
picture cycle. The flicker blade on the 
projector shutter (considering a two- 
bladed shutter) gives a dark period 
lasting 12.5% to be followed by a light 
period of the same, then a long dark 
period consuming 62.5% for pull- 
down and eclipse to permit the other 
eye to see its image (Fig. 29). If stand- 
ard sound-film speed of 24 frames/sec is 
used, the resulting flicker is very an- 
noying. Stepping up the projection to 
48 frames/sec increases flicker frequency 
twice, but it still is noticeable. There is 
a physiological effect that is likely to 
become disagreeably apparent usually 



headache or nausea after a few min- 
utes of viewing pictures projected in this 
way. A complete period of darkness 
for one eye, while light reaches the other, 
will probably always result in visual 
fatigue, if not in nausea, no matter 
how high, within workable limits, the 
flicker frequency is brought. 

Flicker of low frequency calls for 
traction on the control muscles of the 
irises when bright light enters one or 
both eyes. The rapid occurrence of the 
transmission of stimuli, first from one eye, 
then from the other, and the motor 
messages from the brain to the muscles, 
delivered in rapid sequence, probably 
accounts, in part, for the visual discomfort 
experienced by most people when view- 
ing "eclipse" stereo movies. 

Perception of flicker depends upon 
the intensity of the interrupted light, as 
well as the flicker frequency. The 
more intense the light, the higher the 
frequency must go before flicker fusion 
is attained. Also, the larger the angular 
field over which flicker is distributed, 
the greater the consciousness of flicker. 
Hence the dimmer the picture and the 
smaller it is, the lower becomes the flicker 
fusion frequency. 



292 



March 1953 Journal of the SMPTE Vol. 60 




AXIS OF 
POLARIZATION 

Fig. 30. Diagram of the "rotating polarizer" method of 
alternate frame stereoscopic projection. 



There are two ways to project and view 
eclipse stereograms. One is by using 
rotating or vibrating shutter devices held 
up in front of the eyes. These are syn- 
chronized electrically with the projector. 
The other method is to employ a 
rotating polarizer in front of the pro- 
jector lens and polarizing spectacles for 
the viewer (Fig. 30). In one position 
the polarizer delivers light through the 
left spectacle filter, in the other through 
the right filter. 

Alternate-frame, or eclipse, projection 
should have at least twice the number of 
frames required for conventional films. 
That means doubling the length and 
providing for faster projection speed. 
If the alternate frames are photographed 
alternately, there is a very objectionable 
fringing in pictures of moving objects. 
This is a cause of eyestrain, especially 
in a picture where the action seen by 
one eye is in quite a different stage of 
progress than the action seen by the 
other. Difficulty in fusion invariably 
results. This combination of disturbing 
effects, caused by flickers out of phase 
between the eyes and by fusion trouble, 
limits the appreciation of the eclipse 
method. 

Complete visual comfort can be at- 
tained in stereo movies only if the two 
images are projected simultaneously, if 
they are rock-steady, if they are of equal 
brightness, if they are of equal contrast, 



if they are properly aligned vertically 
and horizontally, if far distant points are 
not separated too far in one image from 
that of the other, and if they are of exactly 
the same size. 

Improved Single-Film Methods 

Several inventions by the author 
eliminate the drawback of the single- 
film dual-image arrangement, namely 
the unequal distribution of light. These 
methods employ novel optical systems 
which are accessory attachments to 
standard projector heads. 

If one member of a stereogram has 
even a slightly different brightness, 
some eyestrain will result. If the dif- 
ference is large, the resulting eyestrain 
will be great. This is because the action 
of the eyes' iris diaphragms is entirely 
automatic and not by voluntary control. 
If strong light falls on one eye, both 
pupils will contract and the eye seeing 
the darker image will have its pupil 
closed down more than it should be 
properly to see the image. In accom- 
modation too, both eyes act together 
and it is impossible for one independently 
to accommodate to a different extent 
from the other. 

The improved single-film methods 
satisfy one of the basic requirements for 
good stereograms, easy to look at, in 
that both members of the pair are of 
the same brightness. The illustration 



J, A, Norling: The Stereoscopic Art A Reprint 



293 




Fig. 31. Illustrating the dis- 
tribution of light over a pair of 
images. Left, as in an arrange- 
ment wherein both images have 
the same position. Right, as in 
an arrangement wherein one 
image is "flipped over" in rela- 
tion to the other. 




Fig. 32. Diagram of an improved optical system for simultaneous projec- 
tion of the two images comprising the stereoscopic pair. 



shows a pair of pictures having image 
attitudes in a conventional arrangement. 
The diagram on the right shows an 
improved arrangement. It affords the 
best possible distribution of the light from 
the projector arc (Fig. 31). 

The illuminating spot from the arc is 
considerably more intense in its center 
than in its outer regions. With images 
having the attitudes shown in the left- 
side diagram, more light falls at the top 
of one picture than at the top of the other, 
resulting in a different level of illumina- 
tion in an area of one than in the cor- 
responding area of the other. The ar- 
rangement shown on the right provides 



equal illumination in corresponding areas 
because all portions of the light spot fall 
upon the same corresponding areas in 
each member. This meets the require- 
ment that there be equal brightness for 
each image in order to attain complete 
visual comfort. It may seem surprising 
that the intensity difference between 
center and edges of the light circle is 
significant. Actually, in practice, pro- 
jection engineers are quite happy if there 
is only a 40% loss 60% as much light 
at the edges as in the center. In many 
theater installations, the fall-off is as 
much as 50%. 



294 



March 1953 Journal of the SMPTE VoL 60 



Fig. 33. Diagram of transfer 
of images from the two nega- 
tive films to the single film 
carrying the two disparate 
images. 




x RIGHT EYE 
POSITIVE 



The projection end of this system, 
using standard projector heads, has two 
optical trains containing prisms as well 
as lenses (Fig. 32). These components 
are arranged so that the light beam enters 
and exits normal to the prism surfaces 
and there is no displacement or distor- 
tion such as would take place with wedge 
prisms. Alignment of the images on 
the screen is effected by a micrometer 
control to shift the lenses. In the lower 
optical train, transmitting the right-eye 
image (the one having the conventional 
attitude), the image passes through in 
the conventional manner, reversing to 
"heads up" in the projection lens. 
In the upper optical train, transmitting 
the left-eye image (the one having the 
"flipped over" attitude), the image is 
brought to conventional attitude in a 
pentaprism and right-angle prism for- 
ward of the lens. The polarizers are 
placed in front of the lenses away from 
the intense heat. 

Prints for this system of projection are 
not made directly from the original 



negatives. Duplicate negatives pro- 
duced in an optical duplicating process 
are used instead (Fig. 33). 

Another improved single-film method 
has the images turned on their sides 
(Fig. 34). This arrangement provides 
for the use of the full standard aperture 
in case the pair of images is to occupy 
only one standard frame. A series of 
prism elements between the film and the 
lens turns these images 90. The in- 
verted image is "flipped over" in the 
pentaprism and right-angle prism in 
front of the lens. Otherwise, the prac- 
tical advantages of this variation are 
comparable to those of the method which 
is discussed in the preceding few para- 
graphs. 

Another method, devised by the author 
for attaining equal illumination for both 
members of the stereoscopic pair is also 
illustrated (Fig. 35). Here, there are 
two mirrors arranged between the light 
source and the film. The lower mirror 
is a transmission-reflection type, with a 
reflective coating that will reflect half 



J. A. Norling: The Stereoscopic Art A Reprint 



295 




Fig. 34. Improved single-film method with the images turned on their sides. 




FIRST 

SURFACE 

MIRROR 





50% REFLEC- 
TION -50% 
TRANSMISSION 
MIRROR 



LEFT EYE 






IMAGE 


. . 


\ 




L = = ^ 


a POLAR 






IZERS 


RT. EYE 






IMAGE 




| 



LENSES 



5O-57 



Fig. 35. Illustrating an arrangement that provides equal illumination 
for both images by using a partially transmitting mirror. 



the light to the upper mirror, one-half 
the light passing through to one member 
of the pair. The upper mirror reflects 
one-half the original light beam to the 
other member. Again, two projection 
lenses are used, as in the other systems 
with the polarizers in front of them. 

In this system the images do not have 
to be inverted in relation to each other, 
since the illuminating circle falls upon 
the same corresponding area in each. 
The light-circle can be reduced in size 
to fit the area occupied by one frame, 
adding efficiency. Added efficiency also 
results from the reduction in the number 
of glass-to-air surfaces. Additionally, 
there is less possibility for dirt or "fog- 
ging" on the optical elements. In 
contradistinction to the other methods, 
which require no lamphouse changes, this 
method does require a few changes in 



the lamphouse to accommodate the 
two mirrors. 

A dual-image single-film does not add 
significantly to print costs, to the costs of 
studio production, to shipping and han- 
dling costs between exchange and theater, 
nor does it require dual projectors and 
added labor costs for projection. This 
single-projection method does not intro- 
duce the unwanted possibility of images 
out of synchronism with each other, a 
hazard that exists in dual projection. 
It assures precise registration, one image 
with the other. Alignment is no prob- 
lem, since both images are framed to- 
gether by the same frame-setting lever. 
There can be no jiggling or jittering 
between the images such as is present in 
any dual system no matter how well 
made the projector mechanisms. 



296 



March 1953 Journal of the SMPTE Vol. 60 



OTHER STEREOSCOPIC METHODS 



The Friese-Greene Process 

The first stereoscopic motion picture 
was made by William Friese-Greene 
who patented his process in 1893. 
He used two negative films, one behind 
each lens. The positive images were 
projected side by side on a screen and 
viewed through a cumbersome stereo- 
scope permitting each eye to see only 
the picture intended for it. The com- 
plexity of this system barred it from any 
commercial application. 

The Grid System 

The grid system has been frequently 
proposed and a large number of varia- 
tions on the basic method have been 
suggested during the past 40 years 
(Fig. 36). 

RlULULffLRL R L 




LEFT EYE 



RIGHT EYE 



Fig. 36. Diagram of the fundamental 
"grid system" of stereoscopy. 

Basically, the grid system employs a 
screen containing a large number of 
vertically placed parallel opaque bars 
forming a grating having open or trans- 
parent spaces between them. This 
grid is placed some distance in front of 
the projection screen, the grating in the 
grid being designed so that the right 
eye sees only that portion of the screen 
on which the picture record for the right 



eye appears, the bars in the grating hid- 
ing the left-eye image from the right eye. 
It does the same thing for the other eye. 

The chief problem in using the grid 
system is that the observer's viewing 
distance, angle of view, and of eye 
placement in relation to the grid are of a 
fixed interlocking relationship. Disturb 
one of the three and proper viewing fails. 
A slight shift to the right or left results in 
the breakdown of "correct individual 
seeing" for each eye, and a double image 
becomes apparent, or else a pseudoscopic 
effect results. Improvements have been 
made on the basic grid system, but serious 
shortcomings still remain as an intrinsic 
part thereof. This particularly applies 
to the loss of light. While a light loss is 
common to all stereo projection systems, 
it is particularly severe in this one, due 
to the fact that the opaque areas in the 
grating have to be (in most cases) about 
three times that of the open areas in 
order to keep the images from over- 
lapping. 

Another matter that must be given 
consideration is the appearance of the 
grating to the observer. The dark bars 
and light spaces should be small enough 
to be virtually invisible as a banded 
pattern. To make the grating lines 
invisible, the spacing of the elements 
should be no larger than about ^Vo of 
the viewing distance. 

We shall call the grating space G 8 ; 
then 

G - = 3To (ormore) 

where D v is the viewing distance. 

Thus, if D v = 60 in., G 8 = 0.0175 in., 
approx. ; and if D v = 120 in., G = 
0.035 in., etc. 

The distance, d g , of any selected grat- 
ing in front of the screen depends on the 
relationship of the distance D v to d g , 
or by the relationship of I to D v . 

With the usual grid system, the picture 
through the grating is viewed by con- 
verging the eyes at or near the grating, 



J. A. Norling: The Stereoscopic Art A Reprint 



297 



forming an angle. In order for the 
grating effectively to select the images 
properly for the eyes, the disparate 
members of the stereoscopic pair must be 
projected through the grating at the 
same angle. 

The New York Times, early in 1944, 
reported that Simyon Ivanov, a Soviet 
poster artist, had invented a screen 
made up of tiny squares of thousands 
of strands of fine wire which produced 
3-D effects without the use of eye- 
glasses. James Aldridge, writing on the 
same subject for the North American 
Newspaper Alliance, reported that the 
images reproduced through the screen 
were "coarse and blurred." 

The article goes on to say that the 
original Ivanov grid system had been 
improved and developed in the Sgvin- 
torgkino Studio to the point where it 
promised to become practicable. To 
quote from the article : 

"The glass screen is engraved with more 
than 2,000 converging lines, and it is in 
these markings that the secret of the new 
screen lies. In photographing third-di- 
mensional movies, the only alteration re- 
quired on standard cameras is the addition 
of two or more mirrors fitted near the lenses 
to reflect the images onto the film." 

Obviously, this appears to describe 
Ivanov's photographic process as being 
an application of the beam-splitter 
principle. The article goes on to say : 

"In showing the film, it is projected onto 
'two or more mirrors,' instead of directly 
onto the screen, which reflects the shadows 
onto the glass screen. In turn, the lines on 
the screen unscramble the images, resulting 
in a clearer image than has hitherto been 
obtained in third-dimensional film experi- 
ments." 

Then, in October 1945, the Wall 
Street Journal reported further on the 
Ivanov development: 

"Moscow (AP) The Soviet film indus- 
try is preparing a surprise for the world's 
movie fans a special production of 
Robinson Crusoe to be exhibited on a new 



stereoscopic screen designed to give 
rounded, three-dimensional images. 

"Semeon Pavlovich Ivanov, the inven- 
tor, said that the screen creates an illusion 
so perfect that people unconsciously dodge 
when pictures of birds or airplanes are 
shown. 

"Ivanov said he believed the screen sur- 
passes anything Hollywood had done to 
achieve realism in the exhibition of motion 
pictures." 

On April 29, 1948, the New York 
Herald-Tribune published the following: 

"Moscow, April 28 (AP) The Com- 
munist newspaper 'Pravda' disclosed to- 
day that Semyon P. Ivanov, described as 
the inventor of three-dimensional motion 
pictures, had been removed from the job of 
scientific chief of the special studio in 
which he perfected the invention. 

"The newspaper (Pravda) said that I. 
Bolshakov, Motion Picture Minister, did 
not take Mr. Ivanov's work seriously, tried 
to picture him as a faker and publicity 
seeker and finally pulled him off with the 
excuse that he was freeing him from his 
administrative duties." 

Pravda went on to state : 

"That's how the cinema industry freed 
itself of the worrisome individual whose 
name will go down in the history of the 
Soviet and world cinema." 

Truly, workers on the grid method of 
stereoscopy have a bad time when some- 
one eventually discovers that the images 
don't appear satisfactory throughout 
an auditorium. 

Parallax Stereograms 

Parallax Stereograms are of two prin- 
cipal kinds, one using a grating, or "grid," 
as the selecting screen placed in front 
of the images, and the other using a 
selective screen consisting of small 
cylindrical lens elements, lenticules, 
side by side, and running vertically. 

The "grid" system was introduced by 
Berthier in 1896 and was the first form 
of stereoscopic viewing of still pictures 
that did not require accessories. The 
grid consisted of vertical bars with spaces 



298 



March 1953 Journal of the SMPTE VoL 60 



LEFT LENS , . 
DIAPHRAGM (t) 




GOFFERED 
FILM 



Fig. 37. Above, diagram of one 
method of lenticular stereoscopy em- 
ploying the beam-splitter principle 
with the addition of slit diaphragms. 

Right, diagram of one type of 
camera for making lenticular stereo- 
grams. 






EMUL- 
SION 



FILM 



FOCUSl 



FILM \ 
^ PLANE N 
ALWAYS 
PARAL 

WITH 



ARC THROUGH ^\y^^ 
WHICH CAMERA &/\ I 



LENS SWING 




GOFFERED FILM 



x y 



IMAGE 
PLANE 



in between. The grid was usually made 
on a high-contrast photographic plate. 
The picture was a composite which had 
the two images broken up into bands, 
the image bands for one eye being inter- 
laced between the image bands for the 
other. When the grid was spaced at the 
prescribed distance in front of the com- 
posite and viewed from the correct 
distance, the observer was able to sec a 
binocular view. 

The "lenticular" system used a selec- 
tive screen sometimes registered in front 



of the images; sometimes directly 
bonded to the composite photograph. 
The lenticules do not cut down the re- 
flected light as do the bars in the grid 
system. In one system, the composite 
picture is made in a single-lens camera 
which swings through an arc during 
exposure. The center of the arc is in the 
plane of the subject (Fig. 37). 

Lenticular systems are credited to 
several inventors; among them F. E. 
Ives and his son, Dr. H. E. Ives. Dr. 
Ives called his the "Parallax Panoram- 



J. A. Norling: The Stereoscopic Art A Reprint 



299 



agram." Improvements have been 
made and further developments carried 
out by the Americans, Vanbenschoeten 
and Winnek, and the Frenchman Bonnet. 

The Zafiropulo Process 

Another invention employing the 
lenticular principle is that of Jean 
Zafiropulo, who set out to apply it to the 
motion picture in particular. The proc- 
ess requires extremely accurate align- 
ment of all elements in photography and 
projection. 

The Zafiropulo process involves the 
use of a film containing embossed spheri- 
cal lens elements in its base. Prints must 
have their lens elements exactly aligned 
with those in the negative. The lens 
elements must register with the greatest 
exactitude in relation to the sprocket 
holes in the film. Sprocket teeth, en- 
gaging the sprocket holes, serve as the 
basic registration points for picture 
steadiness. Film shrinkage, which is 
over 0.25% in the lowest-shrink film base, 
will have to be overcome to prevent lens- 
element misalignment with relation to 
the sprocket teeth, and in turn to pre- 
vent misalignment of the film lens- 
elements and the screen lens-elements. 

The Zafiropulo process requires only 
one lens in photography, but it must be 
of large diameter, from 2^ in. to 3 in. 
and such a lens must be of long focus, 
over 6 in. in an //2.5 lens. This is 
about 153 mm compared with 40-mm 
and 50-mm lenses, which are the most 
frequently used in film studios. This 
long-focus lens requires more studio 
space than is needed in conventional 
filming. 

Several other methods have been pro- 
posed for the lenticulated (goffered) 
film process of movie stereoscopy. One 
is, in essence, an application of the "beam 
splitter" principle, differing from it in 
that it produces a series of bands for each 
image (Fig. 37). One band, for part of 
the left-eye image, is formed through 
each lenticule, and another band is 
formed adjacent to it for the correspond- 



ing part of the right-eye image, through 
the same lenticule. The lens must be 
operated at its widest aperture. 

Lip p man's "Integral System" of 
Stereoscopy 

The process of "integral" photog- 
raphy discovered by Lippman in 1908 
utilized a screen composed of an almost 
infinite number of small "lens elements" 
in the form of pin-holes. It affords 
what most who have seen it consider 
the ultimate in stereoscopic viewing. 
It differs from any other system of stere- 
oscopy in that it provides a much larger 
number of images in the plane of the 
photograph and "reduces the number 
of viewing instruments to zero." But 
Lippman's integral photographs can be 
made only as transparencies, and they 
cannot be projected, nor can they be 
reproduced to supply copies that have 
the qualities existing in the original. 

These integrated stereograms are made 
through a screen having a great number 
of pin-holes, each acting as a camera 
lens. No camera is used; the pin- 
holes serving as lenses. The screen can 
be a photographic image in a contrasty 
emulsion on the front side of a glass plate 
with the photographic image produced 
through it on an emulsion on the back. 
The holes must be quite small; their 
size being established by the rules apply- 
ing to pin-hole photography. There 
must be a great number of holes for 
every square inch, and the plates should 
be quite large, 8 X 1 in. or more. 

Exposure is made through the pin- 
holes, and since the effective aperture 
of each pin-hole is extremely small, 
long time exposures are essential. They 
cannot be satisfactorily reproduced; 
hence copies are not obtainable, and the 
negative image has to be rendered into a 
positive by reversal. Viewing should be 
with a mirror placed so that the proper 
left-right attitudes of the images can be 
obtained. 

The nature of the process excludes it 
from practical usefulness, particularly 



300 



March 1953 Journal of the SMPTE Vol. 60 



for motion pictures, but it is an interest- 
ing thing with which to play, and can be 
experimented upon by anyone having 
the required facilities. 

Single-Lens Camera for Motion 
Picture Stereoscopy 

If two strip stereo films are used, such 
things as titles can be made by exposing 
one image, for the required footage, 
then shifting the camera to expose the 
other, as is done in making "slide- 
board" stills referred to earlier in this 
article. 

A variation of the principle can be 
applied to obtain stereoscopic motion 
pictures, especially aerial shots. The 
author has made such films using only 
one camera, and making only one nega- 
tive. Two prints were made from these 
single negatives and projected in inter- 
lock on two machines. However, the 
prints were projected with one print 
having its frames displaced in relation 
to those of the other. The number of 
frames displaced is governed by the 
plane's elevation above the ground and 
above the nearest object, in scenes 
looking straight down. Frame dis- 
placement is also governed by the ground 
speed of the plane. A plane flying at 
100 mph will cover 144 fps, which means 
that a distance of 6 ft is covered in every 
second for a film speed of 24 frames/sec. 
A full reel was made during flights over 
New York City, the camera pointing 
straight down. The plane was flown at 
2500 ft, and slow-motion photography 
of 96 frames/sec was used. 

A film speed of 96 frames/sec with a 
plane speed of 100 mph gives \\ feet of 
advance along the course as registered 
by every frame. Using the equation, 

-4- 50, established the base of the 
D d 

interaxial that would result in the most 
startling visual effect with the least 
eyestrain. The farthest plane, D, the 
ground; the nearest plane, d, the top 
of the Empire State building; and the 
divisor, 50, indicated an interaxial base 



of 40 ft and this was achieved syntheti- 
cally by a displacement of one film with 
the other of 30 frames. Actually, a dis- 
placement of 15 frames was finally 
selected to give the most satisfactory 
results. 

Some stereo shots were made up in this 
way from stock footage taken from a 
plane flying over the Andes. The 
plane was flying about 1 80 mph and the 
exposures were made at normal camera 
speed. The camera pointed horizontally 
toward the distant mountains. 

The most startling of these shots was 
one that included "The Christ of the 
Andes." This heroic statue was in the 
middle distance and stood out in vivid 
relief against the mountains beyond. 

There is one thing that creates quite 
a problem: any unsteadiness in the air- 
plane's flight. This comes out in the 
projection as a constantly changing verti- 
cal and horizontal, and sometimes 
rotational displacement between the 
images. Such displacements existed in 
the films we made and had to be elimi- 
nated by optical duping methods in- 
volving a complexity of steps. But it 
was an interesting experiment and worth 
the trouble. 

The Vectograph 

Vectograph is the name applied to a 
clear plastic sheet on which an image 
may be rendered in terms of varying 
degrees of polarization, and viewed 
through a polarizing filter. The vecto- 
graph can accommodate an image on 
both its sides, and each image can be 
made to have its axis of polarization at 
right angles to that of the other. A 
stereo vectograph has the images of a 
stereo pair printed respectively on top 
and bottom of a vectograph sheet, and 
is viewed through polarizing spectacles 
with its respective windows having polar- 
izing axes corresponding to those of the 
vectograph images. In slide form, the 
3-D vectograph can be shown in a stand- 
ard monocular projector without filters. 
As in other systems using polarized light, 



J. A. Norling: The Stereoscopic Art A Reprint 



301 



silver non-depolarizing screens and polar- 
izing viewing spectacles are required. 

During World War II the 3-D vecto- 
graph was used by the armed services 
for aerial reconnaissance and for training 
personnel in various skills. Other uses 
for this novel, paper-thin stereo picture 
will doubtless be found. Its picture 
quality is excellent and its ease of han- 
dling and processing are distinct advan- 
tages for any photographic process. 

3-D Pictures and Television 

Inevitably today, any new method of 
visual presentation, both still and motion, 
can be telecast. As a matter of fact, 
an experimental stereo-television system 
has been at work in the Argonne National 
Laboratories. It permits an operator to 
keep a precise watch over the "hot" 
materials he is handling by remote 
control. Equipped with a pair of 
viewing spectacles and with eyes glued 
to a pair of television screen images 



which are transmitted by a binocular 
television camera, the operator does get a 
3-D impression. But the problems in- 
volved in presenting satisfactory 3-D 
television to the public may be so very 
great that, by comparison, the problems 
of introducing color television have been 
small indeed. It is anybody's guess 
when stereo television will enjoy a wide- 
spread audience ; if the history of stereo 
movies can provide a clue, the time is 
many years away. 

Nonstereo System 

A system which has recently been 
described as producing a 3-D effect is 
the Cinerama development of Fred 
Waller. This method, requiring a 
multiplicity of cameras and projectors, 
presents a dramatic panoramic view of 
the scene photographed. But it is not 
3-D at all, since it does not present a 
mutually exclusive image to each eye, 
the basic requirement of any 3-D system. 



GENERAL DATA 



For successful stereoscopic projection, 
it is important to know the size to which 
the picture is to be projected, for upon 
this knowledge depends comfortable 
viewing or maximum effectiveness, or 
both. Distortion of the subject for the 
spectator depends on his viewing dis- 
tance and the angle at which he views the 
picture, but as in conventional movies, 
distortion due to the spectator's view- 
point is not serious unless he is very far 
off screen center or extremely close to or 
very far away from the screen. 

Stereoscopic Terms 

There are certain terms used in stere- 
oscopy. These are : 

The Stereoscopic Window: The "frame" 
behind which the scene apparently exists 
in space. In some cases, elements of 
the scene may extend in front of the win- 
dow. These are special effects, and do 
not belong in a discussion of basic 
principles. The symbol is Sw. 



Interocular (I n ) : The distance apart 
of the human eyes. We may select 2.5 
in. as the interocular. 

Inter axial (I): The distance apart, 
or horizontal spacing of the stereo 
camera lenses, more truly the spacing 
apart of the central axes of the picture 
area. 

Major Distance (D): The distance 
from the lenses to the farthest object. 

Minor Distance (d): The distance 
from the lenses to the closest object. 
This is often the plane at which the 
stereoscopic window is planned to exist. 
In most cases, the nearest obiect will be 
a slight distance in back of the nearest 
plane. 

Width of Image (w) : Refers to the width 
of that part of the negative which is to 
be used in the final prints. For standard 
35mm theatrical film, this is 0.825 in.; 
for 16mm, 0.4 in.; for Stereo-Realist 
slides, 0.875 in. 

Focal Distance (F): Refers to the dis- 



302 



March 1953 Journal of the SMPTE Vol.60 



tance from the principal node of the lens 
to the film, and does not refer to the 
equivalent focal length of the lens, 
although in most cases the stated focal 
length may be considered the focal dis- 
tance. It is only for extreme close-ups 
that a differentiation must be made. 

Parallax Index (P): This refers to the 
parallactic difference between disparate 
members of the stereoscopic pair when 
projected. The parallactic difference 
is determined by the following factors: 

1. The distance to the nearest plane 
in the subject. 

2. The distance to the farthest plane 
in the subject. 

3. The ratio between image width 
and focal distance of the lenses used. 

4. The interaxial used. 

Parallax index can be expressed by 
the equation 

W T\A 

I = 



PFD-d 

This equation was derived irom the ac- 
companying diagram, which represents 
the geometry of the system of taking 
stereoscopic photographs (Fig. 38). 
Calculations are simplified if we consider 
one of the lenses collinear with the far 
and near points. 

Points D and E represent the positions 
of the two lenses. The distance DE is 
the interaxial "I." Distance DA or EA 
is the lens to film distance, and its symbol 
is "F." F and G are points on the near- 
est and farthest planes in the subject. 
DF is "d" and DG is "D." BC is the 
horizontal shift on the image plane 
created by points F and G, and this 
distance has the symbol "S." 

It will be seen that, if the nearest point 
is superposed on the screen, the far- 
thest point will be separated by an 
amount determined by the interaxial 
used in taking the picture. The maxi- 
mum value of the distance of homologous 
points on the screen should not exceed 
2\ in. to avoid visual discomfort. 
To the observer, homologous points 
back of the stereoscopic window which 




Fig. 38. The geometry of the 
stereoscopic process. 

are 2\ in. apart should appear to be at 
virtual infinity. Some individuals can 
tolerate 3j to 4 in., but it is best not 
to use "super infinity" if it can be avoided. 
Homologous points that superpose on the 
screen will seem to lie in the plane of 
the stereoscopic window. 

The minimum distance at which ob- 
servers can look at a picture comfortably 
when it has a 2^-in. maximum sepa- 
ration of points is about 6 ft. In general, 
the maximum horizontal displacement 
should not exceed ^ of the viewing 
distance. This requirement exists be- 
cause in viewing projected stereo pictures, 
the observer must "uncouple" the facili- 
ties of convergence and accommodation 
of the eyes. 

In viewing, it is desirable that the 
parallax index be 24 or greater. 

Parallax index P may be stated as 



where D v is the viewing distance, and 
In is 2.5 in. 

Thus where D T = 60 in., P = 24; 
where D v = 120 in., P = 48, etc. 



J. A. Norling: The Stereoscopic Art A Reprint 



303 






Fig. 39. The principle of lens angle 
and viewing angle. 



Fig. 40a. Representation of a cluster 
of light vibrations. 



General Rides: 1 . A general rule can be 
laid down for the photography and pro- 
jection of stereoscopic views: The pro- 
jected view should have the same angular 
dimension for the viewer as the scene taken 
by the camera. This is the ideal, but never 
attainable in practice except for just 
one person at the prescribed distance 
from the screen and viewing the screen 
along the projection axis (Fig. 39). 

2. The apparent depth of the stereo- 
scopic view should be the same as the 
real depth of the scene. To attain the 
right apparent depth, the correct inter- 
axial must be employed. The required 
interaxial varies over a wide range and 
for projected views must be given much 
more serious consideration than for 
hand-held views. If we plan to project 
on an 18-ft wide screen, we must not 
use as wide an interaxial as we can use 
on a 6-ft screen, because if we do, and 
employ convergence to establish the 
stereoscopic windows, we are either going 
to have homologous points at infinity 
spread so far apart that the eyes have to 
diverge to accommodate for them, or we 
are going to have to adjust the projected 
stereoscopic window to a plane far in 
front of the screen. 

It is not difficult to arrive at the cor- 



rect interocular if we use the simple 
equation 

wed 

I (interaxial) = 
sf 

where w = width of the image on the 

film, 
e = normal human interocular 

(2|in.) 
d = distance from the camera 

lens to a plane just in 

front of the nearest object 

(plane of convergence), 
s = width of the projected 

picture, 
f = focal length of the camera 

lenses. 

The establishment of the stereoscopic 
window is not of great importance in 
hand-held views, but it must be em- 
ployed in projection, and properly em- 
ployed. If it is not, there will be the 
marginal disturbances that have been 
mentioned before, and they are hard 
to look at. There is nothing in natural 
vision to correspond to them, and since 
the ideal stereoscopic view is one that 
should afford complete visual comfort, 
the appropriate window frame should 
be calculated in every scene. If it is 
not, people may have trouble looking 
at your stereoscopic "masterpieces." 



304 



March 1953 Journal of the SMPTE VoL 60 




Fig. 40b. Light passing through two plane polarizers: A, with axes parallel; 
B, with axes at 90 to each other; C, with axes at less than 90. 



Very striking stunt shots can be made 
stereoscopically. For instance, objects 
can be made seemingly to float in space 
between screen and observer provided 
the object is well inside the margins of 
the picture areas. Objects should not 
be photographed so as to appear so 
near to the person observing the pro- 
jected images that he will have trouble 
fusing them. Consideration must be 
given to the accommodation limits of 
the eyes; that is, for convergence ac- 
commodation limits. 

Theoretical accommodation limits of 
the human eye in convergence are based 
on normal close reading distance of 
15 in. (Note: This formula does not 
take into consideration what physiologi- 
cal effect, if any, is introduced when the 
accommodation muscles are used with- 
out correlative focusing.) The angle 
of convergence of the eyes (interpupilary 
distance of 2.625 is used) at a distance 
of 16| in. is slightly more than 9. 
The displacement of the disparate images 
on the screen are given in inches and 
decimals of an inch. 

Formula: Observer distance from 
screen less 15 in. multiplied by the 



tangent of half the angle subtended by 
the eyes gives the maximum separation 
on the screen at a given distance. To 
obtain the required separation of images 
on the film, divide the projection aper- 
ture width by the screen width and mul- 
tiply by the separation of the projected 
images (tan 4 30 ft X distance from the 
screen). 



Observer 

distance from 

screen, ft 

4 

6 

8 

10 
12 
16 
20 
24 
30 



Maximum 

separation 

on screen, in. 

2.60 

4.48 

6.37 

8.26 

10.15 

13.93 

17.71 

21.48 

27.15 



An Analysis of Light Polarization 

Since the phenomenon of light polar- 
ization is so closely related to the prac- 
tice of stereoscopy, it is of benefit briefly 
to review it. 

Let us imagine we are looking head-on 



J. A. Norling: The Stereoscopic Art A Reprint 



305 




Fig. 41. Vector diagram of a vibration. 

at a beam of light and that we can con- 
ceive it as a bundle of rapidly vibrating 
arrows pointing outward in an infinite 
number of directions (Fig. 40a). A 
polarizing filter can cause all vibrations 
to take place parallel to each other 
(Fig. 40b). 

The polarizer transmits not only the 
vibrations which are originally parallel 
to the polarization axis, but all the com- 
ponents of all the infinite number of 
vibrations at angles to the axis. The 
amplitude of any vibration along the 
axis is equal to 

At cos a 

where a is the angle between the direc- 
tion of vibration and the axis (Fig. 41). 
Since the energy of a vibration is 
proportional to the square of its ampli- 
tude, the relative intensity I u of light 
transmitted by two polarizers with axes 
at any angle from to 90 is given by 

I u = I cos 2 a 

when the angle between the polarizing 
axis is a and I is the relative intensity 
of the transmitted light when the angle 
a is zero. 

A graphical representation of this, with 
I arbitrarily equal to unity, shows the 
relative intensity of the light through 
two polarizers with axes at various angles 
to each other (Fig. 42). This curve 
is true only for perfect polarizers which 
would have a transmission of 50%. 



1.00 
0.80 
0.60 
0.40 
0.20 
0.00 



0* 10* 20* 30*40 50* 60* 70* 80* 90 

Fig. 42. Relative transmission of light 
through two polarizers arranged with 
axes at angles to each other, from 
to 90. 

Actually, the best polarizers have a 
transmission of only about 40%. 

Conclusion 

The fundamental requirement of any 
stereoscopic system is that each eye sees 
only that member of the stereoscopic 
pair intended for it and excludes the 
image belonging to the other eye. 
The disparate images of the stereo- 
scopic pair must be distributed to the 
eyes of the audience in a selective man- 
ner. To quote from Dr. H. E. Ives: 

"There are only two places where 
the distribution of images to eyes 
can be done ; these are at the screen and 
at the eyes. The number of images 
at the screen can be reduced to two, if 
the number of viewing instruments is 
equal to the number of spectators. 
The number of viewing instruments can 
be reduced to zero if the number of 
images at the screen is made infinite. 
Any gain in simplification at one point 
is offset by increase in the complexity 
or expense at the other." 

Editor's Note 

When this paper was reviewed for 
reprinting, it was approved by the Board 
of Editors, with two recommendations: 
(1) that the illustrations be given numbers 
and referred to in the text; and (2) that 
references be prepared, keyed to the 
respective portions of Mr. Norling's text. 



306 



March 1953 Journal of the SMPTE VoL 60 



Except for the man with the derby and 
stereo camera (omitted in this reprint of 
the original), the illustrations were easily 
enough numbered; but to prepare and 
key a reference list was more than could 
be done in the time available, so the 
following bibliography is presented. It 
is not a complete bibliography nor is it 
even necessarily a complete guide to all 
the details and bases behind Mr. Norling's 
reporting. This bibliography is offered 
for what it is worth in relation to this 
article. 

A more comprehensive bibliography on 
stereoscopic and wide-screen motion pic- 
tures is expected soon from the Society's 
Stereoscopic Motion Pictures Committee. 

Bibliography 

Bibliography of Stereography. Four hundred 
references published in mimeo form by 
The Stereo Society of America, covering 
magazines and journals from this 
country and from abroad. The ref- 
erences have a wide range from 
editorials and popular articles to learned 
treatises. Copies of the Bibliography 
are available at SI. 50 each from the 
office of the Librarian of the Stereo 
Society, 274 Pearl St., New York 38, 
N.Y. 

Bodrossy, Felix, Magyar Technika, May 
1952. 

Brewster, Sir David, F.R.S., The Stereo- 
scope, John Murray, London, 1856. 

Colardeau, E., Traite general de Stereoscope, 
J. de Francia, Paris, 1923. 

Dewhurst, H., "Auto-precision stereos- 
copy," Phot. J., Sec. B, 92B: 2-24, 
Jan.-Feb. 1952. 

Dudley, L., "Stereoscopy in the Tele- 
cinema and in the future," J. Brit. 
Kinemat. Soc., 78: 172-181, June 1951. 

Eder, Josef Maria, History of Photography, 
Columbia University Press, New York, 
1945, p. 383. 

French Patent No. 938,023. M. Bonnet, 
"Etude du precede de relief 'verivision 
holding,' " Bull, de I'Assoc. Franq. des In- 
genieurs et Techniciens du Cinema, 9: 11-16, 
1951. 

Gillett, A., Chretien, H., and Tedesco, J., 
"The panoramic screen projection equip- 
ment used at the Palace of Light at the 
International Exposition (Paris, 1937), 
Jour. SMPE, 32: 530-534, May 1939. 



Hardy, Arthur C., "The depth of field of 
camera lenses with special reference to 
wide film," Jour. SMPE, 16: 286-292, 
Mar. 1931. 

Ives, H. E., Traill-Taylor Lecture before 
the Royal Photographic Society, 1933. 

Judge, A. W., Stereoscopic Photography, 3rd 
rev. ed., Chapman & Hall, Ltd., Lon- 
don, 1950. 

Kaplan, Sam H., "Theory of parallax bar- 
riers," Jour. SMPTE, 59: 11-21, July 
1952. 

Kennedy, Clarence, "The development 
and use of stereo photography for educa- 
tional purposes," Jour. SMPE, 26: 3-17, 
Jan. 1936. 

Kitrosser, Samuel, "Stereoscopic pho- 
tography as applied to vectographs," 
Organic Chemical Research Seminar, 
Polaroid Corp., May 16, 1945. A 
somewhat revised version of this thesis, 
now entitled "The Polaroid Interocular 
Calculator" is scheduled for publication 
in PSA Journal, Sec. B., May 1953. 

Kriebel, R. T., "Stereoscopic photog- 
raphy," The Complete Photographer (Pola- 
roid Corp.), Nos. 51 and 53, 1943. 

Land, Edwin H., "Vectographs: images in 
terms of vectorial inequality and their 
application in three-dimensional repre- 
sentation," J. Opt. Soc. Am., 30: 230, 
June 1940. 

Law, H. B., "A three-gun shadow mask 
color kinescope," Proc. I.R.E., 39: 1186- 
1194, Oct. 1951. 

Law, R. R., "A one-gun shadow mask 
color kinescope," Proc. I.R.E., 39: 1194- 
1201, Oct. 1951. 

McLaren, Norman, "Stereographic anima- 
tion," Jour. SMPTE, 57: 513-520, Dec. 
1951. 

Newcomer, H. Sidney, "Wide screen pho- 
tography with cylindrical anamorphosing 
systems and characteristics of motion 
picture lenses and images," Jour. SMPE, 
20: 31-53, Jan. 1933. 

Norling, J. A., "Three-dimensional motion 
pictures," Jour, SMPE, 33: 612-634, 
Dec. 1939; "Progress in three-dimen- 
sional pictures," Jour. SMPE, 37: 516- 
524, Nov. 1941. 

Report on Screen Brightness Committee 
Theater Survey, Jour. SMPTE, 57: 238- 
246, Sept. 1951. 

Rule, John, T., "The geometry of stereo- 



J. A. Norling: The Stereoscopic Art A Reprint 



307 



scopic projection," J. Opt. Soc. Am. y 31: woode, Nigel, The Theory of Stereoscopic 

325-334, Apr. 1941. Transmission, University of California 

Savoye, Francois, "Les precedes de cinema PresSj Berkeley, Calif., 1953. 

en relief," Rev. photographie optique: p. 23, Szegho? c ^ "Experimental tri-color 

Feb 1951; "Vision direct ," Rev. photo- cathode . ra tube/5 Tde . Tech , 9: 34^35, 

graphic optique: p. 34, Dec. 1950 y 

Segaller, Denis, "The Russian stereoscopic 

cinema," Discovery, 355-358, 361, Nov. Vanfet > Pierre > " La P rise de e et la P r - 

1949 jection en relief en vision collective 

Spottiswoode, Raymond, "Progress in directe," Rev. photographie optique: p. 59, 

three-dimensional films at the Festival of Mar. 1951. 

Britain," Jour. SMPTE, 58: 291-303, Witherow, G., "An experiment in three- 

Apr. 1952. dimensional movies," Home Movies: p. 

Spottiswoode, Raymond, and Spottis- 170, Apr. 1941. 



308 March 1953 Journal of the SMPTE Vol. 60 



I Reels, Magazines, Spindles for 3-D 



For theater owners who are planning to 
equip for 3-D, the present magazine-reel- 
spindle situation is chaotic, to say the least. 
It may therefore, serve a useful purpose 
to discuss a number of factors related to 
the current problem, as viewed by the 
engineers, and to report on efforts now be- 
ing made to smooth the path toward con- 
version. 

Reel Size 

The vast majority of theaters have only 
two or at best three projectors, and since 
3-D today requires simultaneous use of 
two machines, the "continuous per- 
formance" goes out. The length of an 
"average" motion picture program, includ- 
ing feature and some combination of 
abbreviated second feature, newsreel or 
short subjects, will run about two hours, 
as a conservative estimate, and at the 
standard projection rate of 90 ft/min, 
will use 90 X 60 X 2 or 10,800 ft of film. 
As it is physically impractical to put two 
! miles of film on a single reel, theaters must 
* use more than one reel per program. 
Most will need one or more intermissions 
to permit rethreading the two projectors; 
the minimum is obviously preferred. 

Reel size is not critical if more than one 
intermission is planned (although a stand- 
ard will be necessary). Considering the 
single intermission, therefore, it would 
appear that a reel should have a minimum 
capacity of 5400 ft. Assuming a hub 
diameter of 5 in. (presently contemplated), 
it has been estimated by the Motion 
Picture Research Council and others that 
this footage does require a 24-in. diameter 
reel. Exact determination will depend on 
such variables as take-up tension, thick- 
ness of film, condition and cleanliness of 
film, etc. It should be noted that a reel 
can hold roughly 10% more black-and- 
white than color film because of the 
difference in film thickness. 

However, the only large reels now being 
produced in quantity, against orders from 
a large number of exhibitors, are 23 in. 
and not 24 in. The reasons for this are 
twofold: partly economic and partly 
based on the technical considerations of 
the early 3-D productions. 



First: Bwana Devil totals 7600 ft and 
House of Wax reportedly will have a release 
footage of 8200 ft or approximately 10-15% 
less than the average 2-D and expected 
3-D features. Capacity of the 23-in. reel 
is probably in the order of 4500-4800 ft. 
This capacity is therefore adequate for 
running one of these features plus a newsreel 
or short spliced to the end of reel two of the 
feature. 

Second: At this critical moment of 
demand only 24-in. magazines and 23-in. 
reels are available. These were initially 
designed for some of the larger houses who 
wanted to decrease the number of change- 
overs and for the export market where the 
demand stemmed from single-projector 
theaters. To capitalize on the widespread 
initial interest in 3-D, the limited available 
stock was purchased and immediate orders 
placed for future delivery. 

On the one hand, flexibility in film 
production and exhibition appear to call 
for a 24-in. reel and 25-in. magazine; on 
the other hand, immediate usefulness and 
availability of 23-in. reels and 24-in. 
magazines dominate the scene for the 
moment. 

Magazines 

Some exhibitors have voiced concern 
as to the possibility of installing the 24-in. 
or 25-in. magazines on their equipment. 
This has been with particular reference to 
theaters still using Universal bases. There 
is definite information that either modi- 
fication can be made with but few excep- 
tions. The exceptions occur where the 
Universal base is used in very cramped 
quarters. In those cases smaller magazines 
will have to be used and the number of 
intermissions increased. 

Spindle Size 

Diameter of the spindle in present use 
with 2000-ft reels is & in. With the load 
increased some 250% by the larger reels 
of film, it is generally felt that the spindle 
diameter must be increased to withstand 
the increased stress. No definite recom- 
mendations have been made as yet al- 
though in. has been mentioned as a 
likely value. 



309 



Take-up Tension 

With regard to the 24-in. reel, the 
MPRG has stated that "Because of their 
size and weight, such reels must have free 
wheeling flanges to minimize strain on the 
film when the machine is started." By 
making the flanges free-wheeling, the reel 
inertia is appreciably decreased. And, of 
course, it is the flywheel effect of the flange 
mass which produces the film damage; 
that is, the reel tends to stay at rest after 
the projector is started and to continue 
at a fixed speed once it is in motion. 
Therefore, with fast start projectors, slack 
is created at the start and when the reel 
catches up it tends to maintain its speed 
and consequently gives an awful wallop 



to the film perforations engaged in the 
hold-back sprocket. 

Essentially what is required is a method 
of applying an adequate and constant 
take-up tension from the beginning of the 
start-cycle through to the end of the reel. 
Several methods, other than the free- 
wheeling reel, have been mentioned as 
potential solutions: use of a separate- 
drive, constant-torque motor, the double- 
cone type of mechanism and others. 

Committee Meeting 

The Film Projection Practice Committee 
has scheduled a meeting to consider those 
aspects of these questions which lend 
themselves to and require standardization. 
This will be reported in the next Journal. 
Henry Kogel, Staff Engineer. 



73d Convention, Los Angeles Statler, April 27 - May 1 



Stereoscopic motion pictures and the 
engineering of drive-in theaters are phases 
receiving special attention at the Spring 
Convention. The technical and entertain- 
ment program is outlined on cards mailed 
on March 2 to all members in the western 
hemisphere. This card and the Advance 
Program, to be mailed March 27 to all 
members, will be sent to anyone on 
request. 

President Herbert Barnett has announced 
plans for the Get-Together Luncheon: 

Mitchell Wolfson, noted exhibitor, tele- 
vision broadcaster and former President 
of Theatre Owners of America, will speak 
at the opening luncheon, Monday noon. 
His keynote address will bring to Holly- 
wood and to the engineers, the exhibitors' 
views on stereoscopic and wide-screen 
motion pictures, present outlook for 
theater television, and current prospects, 
technical and economic, for the drive-in 
theaters of the United States. 

The Advance Program will give the full 
schedule, with abstracts, of the papers 
that Chairman Ralph Lovell now has 
firmly committed. 

Stereo is scheduled for Monday after- 
noon and evening and Tuesday afternoon. 
Stereoscopic and wide-screen principles 
will be presented, along with photography 
and exhibition demonstrations. A stereo- 



phonic sound demonstration has been 
planned for Friday evening. 

Television begins on Tuesday with a 
papers session and a tour of the new CBS 
Television City. Tuesday evening's ses- 
sion will have a progress report on NTSG 
color standardization. (The April Journal 
will have the popular paper on this subject 
presented at the Washington Convention 
by A. V. Loughren.) Subscription tele- 
vision will also be on the program. The 
television sessions have been scheduled 
to precede the NARTB engineering 
sessions which begin on April 29. 

On Wednesday morning and afternoon 
it is drive-in theaters factors affecting 
picture and sound quality and new equip- 
ment. 

High-speed photographers take a trip 
on Wednesday to the U.S. Naval Ordnance 
Test Station's Morris Dam Test Facility 
at Azusa, Calif. High-speed papers are 
scheduled for Thursday evening and Friday 
morning. 

Wednesday evening is set aside for the 
traditional Cocktail Hour and the Semi- 
annual Banquet and Dance. 

And the morning after the ball it's 
open house at the Hollywood film proc- 
essing laboratories. Thursday afternoon 
offers film processing subjects and Thursday 
night, a general session. Friday's sessions 
are on sound recording and on magnetic 



310 



striping and editing. (All the Wash- 
ington Convention magnetic striping sym- 
posium papers are to be published as Part 
II of the April Journal.} 

There are now ten committee meetings 
and five committee reports scheduled 
during the week. 

The Ladies Program is highlighted by 
an outdoor luncheon by the Huntington 
Hotel pool and tour of the Huntington 
Library in Pasadena, on Tuesday. A 
luncheon and tour of a motion picture 
studio are planned for Thursday and a 
luncheon on Wednesday at the Statler 
will precede a fashion show in the Terrace 
Room. Tickets for outstanding television 
and radio shows will be available at other 
times to help complete an active and 
interesting week. 

There will be an equipment exhibit 
arranged particularly for new equipment 
described in papers on the technical 
program. Questionnaire forms are due 
by April 1 to Thomas J. Gibbons, Exhibit 
Chairman, 446 N. La Brea Ave., Holly- 
wood 36, Calif. 

Opening of the Registration Desk has 
been moved up to Sunday afternoon by 
Convention Vice-President Jack Servies 
and Local Arrangements Chairman 



Vaughn Shaner. The people now making 
the plans and who will carry the Con- 
vention through are: 
Program Chairman Ralph E. Lovell 
Papers Committee Chairman, W. H. 

Rivers Vice-Chairmen, J. E. Aiken, 

Skipwith W. Athey, Geo. W. Colburn, 

G. G. Graham, John H. Waddell 
73d Convention High-Speed Photography 

Carlos H. Elmer 

Local Arrangements Vaughn C. Shaner 
Banquet Sidney P. Solow 
Door Attendants USC Student Chapter 
Exhibits Thomas J. Gibbons 
Hotel Reservations and Transportation 

Philip G. Caldwell 
Luncheon Loren L. Ryder 
Membership and Subscriptions John W. 

Duvall 

Motion Pictures Ted Fogelman 
Projection, 35mm and 16mm Chairman, 

M. B. Smith, assisted by Merle Chamber- 

lin and Allan C. Curtis 
Public Address and Recording Edwin 

W. Templin 

Publicity Harold Desfor 
Registration and Information Robert 

Young 
Ladies Reception and Registration Mrs. 

Vaughn C. Shaner 



The Seventy-Fifth Semiannual Convention 



A year from now the Society will hold its 
seventy-fifth semiannual convention. The 
event will be marked by emphasis on the 
historical aspect of the development of 
motion picture engineering. 

During the Chicago Convention last 
year plans were initiated and John Frayne 
accepted the Chairmanship of a special 
committee formed to accomplish a special 
program which will include authoritative 
historical papers outlining the development 
of the cinematographic arts and associated 
technologies. The Committee is also 
working on the selection of pioneers who 
will be properly honored at this Conven- 
tion for their contribution to the industry. 
The roster of the Committee is : 

John G. Frayne, Chairman 

Edmund A. Bertram 

Frank E. Cahill 



James Card 
John I. Crabtree 
James W. Cummings 
Emery Huse 
Axel G. Jensen 
Barton Kreuzer 
J. B. McCullough 
Peter Mole 
H. W. Moyse 
Terry Ramsaye 
E. I. Sponable 

The Historical and Museum Com- 
mittee, under Chairman John B. Mc- 
Cullough, will contribute substantially to 
the 75th Convention Program. 

Suggestions or proposals for any aspects 
of the 75th Convention will be welcomed 
by John G. Frayne, 6601 Romaine St., 
Los Angeles 38, Calif. V.A. 



311 



1953 Nominations 



The Nominating Committee under the 
Chairmanship of Peter Mole, 941 North 
Sycamore Ave., Hollywood 38, Calif., is 
now considering candidates for election 
to national offices of the Society. The 
Chairman invites all Fellows and Honorary 
and Active Members to submit their sug- 
gestions. At the end of 1953 nine vacan- 
cies are to occur they will be in the 
offices of Engineering Vice-President, 
Financial Vice-President and Treasurer 
and the following Governorships: two 



from the East, two from the Central area 
and two from the West. Incumbents are 
listed on the inside back cover of each 
Journal. 

Mr. Mole requests that suggestions be 
sent to him at the earliest possible date 
because the Committee has a substantial 
amount of work to be accomplished as is 
specified in Bylaw VIII which appeared 
last on page 345 of the Journal for April 
1952. 



Pacific Coast Section Meeting 



The February Pacific Coast Section meet- 
ing with a program entitled "Three- 
Dimensional Motion Pictures" brought out 
an unprecedented attendance of 800. It 
was originally intended to open the meet- 
ing to all interested motion picture and 
television engineers, but the response to 
the first notice was so great that follow-up 
notices had to be sent advising that ad- 
mittance would be restricted to members 
of the SMPTE plus members of the Holly- 
wood Section of the American Society of 
Cinematographers. 

The program, held on the sound stages 
at the Republic Picture Studios, presented 
speakers from the 3-D field and short 
film demonstrations. Sol Lesser Pro- 
ductions provided a Stableford screen from 
England. Members examined it and asked 
questions about it following the meeting. 
Mr. Lesser very kindly provided the 
Polaroid glasses for viewing the films. 

Dr. Harold R. Lutes, President of the 
H-L Instrument Company, and a manu- 
facturer and developer of optical instru- 
ments and photographic equipment, spoke 
briefly on some of the psychological prob- 



lems encountered in stereoscopic photog- 
raphy and illustrated his discussion with 
a series of slides designed to show the 
optical illusions and depth perception 
sensations possible with three-dimensional 
photography. 

Raymond J. Spottiswoode, Technical 
Director of Stereo Techniques, Ltd., 
London, spoke on "Practical Aspects of 
Three -Dimensional Motion Picture Photog- 
raphy," and showed three of Stereo 
Techniques' films, one an animated pro- 
duction. Mr. Spottiswoode's discussion 
was of great interest to the group, very 
well presented and greatly heightened by 
a charming sense of humor and under- 
standing of his audience. 

Raphael G. Wolff, President of Raphael 
G. Wolff Studio and Stereo-Cine, Inc., 
offered film demonstrations from current 
productions to illustrate some of the results 
available today to the professional motion 
picture producer or commercial film 
producer through the use of stereo photog- 
raphy. Philip G. Caldwell, Secretary- 
Treasurer, Pacific Coast Section. 



SMPTE Lapel Pins are available from 
the Society's headquarters. They are 
gold and blue enamel, with a screw back. 
The pin is a |-in. reproduction of the 
Society symbol the film, sprocket and 



television tube which appears on the 
Journal cover. The price of the pin is 
$4.00, including Federal Tax; in New 
York City, add 3% sales tax. 



312 



Book Reviews 



Filter Design Data 

for Communication Engineers 

By J. H. Mole. Published (1952) by 
John Wiley, 440 Fourth Ave., New York 
16. i-xvi + 246 pp. + 4 pp. index. 127 
illus. 56 tables. 6 X 9f in. Price 

$7.50. 

This volume, as its title indicates, is 
primarily intended for the engineer con- 
cerned with filter design. Emphasis is 
on design problems to the almost complete 
exclusion of theoretical considerations. 
Derivations of formulas have been included 
only when necessary to the clarity of pres- 
entation. All filters treated are, with one 
exception, Zobel filters. This exception 
is the coupled resonant bandpass filter. 
An elementary knowledge of transmission 
and filter theory has been assumed. For 
a person with such a background, the 
explanation of design methods is adequate. 

The author has succeeded in presenting 
a large amount of design data in this 
252-page volume. More than 70 of the 
127 figures are charts directly concerned 
with numerical evaluation. Design data 
for a large number of specific filter con- 
figurations (in excess of 30) is presented. 
These vary from a single section to a filter 
of five M-derived sections (with different 
M values) plus a terminating k section. 
Sufficient data are presented to enable one 
to determine the minimum number of 
elements required for the solution of a 
given design problem and to assign element 
values after having determined the required 
configuration . 

Included is a chapter on the determina- 
tion of tolerances required on individual 
elements to meet a specified overall 
tolerance. This problem is approached 
by the theory of probability. G. W. Read, 
Westrex Corp., 6601 Romaine St., Holly- 
wood 38, Calif. 



Comptes rendu Proceedings of 1951 
Stockholm Convention of the Commis- 
sion Internationale De L']2clairage 

Published (1952) by the Central Bureau of 
the^ Commission Internationale De 
L'Eclairage. 1706 pp. Available at a 
delivered price of $12.50 per set of 3 vols., 
from T. D. Wakefield, Treasurer U.S.N.C., 
I.C.I., F. W. Wakefield Co., Vermilion, 
Ohio. 

The twelfth session of the International 
Commission on Illumination was held in 
Stockholm, Sweden, in June and July, 
1951. The proceedings are published in 
three volumes described as follows: 

Volume I contains the Secretariat Re- 
ports for the separate committees partici- 
pating. The twenty-odd contributing 
committees cover such subjects as defini- 
tions and units, methods of measurement, 
physiology and theories of vision, light 
sources and all sorts of practical applica- 
tions of lighting. The Secretariat Reports 
contain much material contributed from 
many different countries which should be 
of great value to any student or research 
worker interested in the subjects covered; 
for, in many instances, they contain very 
extensive bibliographies consisting of liter- 
ally hundreds of references to papers from 
many lands. The report of Committee 
62d on Cinema Lighting contains informa- 
tion submitted by members from Belgium, 
Czechoslovakia, Denmark, France, Great 
Britain, Norway, Sweden and the U.S.A. 
It is gratifying to note that so many coun- 
tries by independent means have arrived 
at approximately the same range of recom- 
mended screen brightness as our own for 
viewing 35mm motion pictures. 

Volume II contains forty-five papers on 
a wide range of subjects presented at the 
Stockholm Meeting by individual dele- 
gates. 



SMPTE Officers and Committees: A new publishing of the roster of Society 
Officers and the Committee Chairmen and Members is scheduled for the April 
Journal. The last one is in April 1952. 



313 



Volume III contains a complete list of 
the delegates and visitors, minutes of the 
meetings, and the official recommenda- 
tions of the C. I. E. Many of the recom- 
mendations of committees of the G. I. E. 
have already been published in various 
journals. Some were contained on pages 
734-738 of the Journal of the Optical Society 
of America, Vol. 41, Oct. 1951. Recom- 
mendations by Committee 62d on Screen 
Lighting in Cinemas and those of Com- 
mittee 63 on Television were reported on 
pages 283 and 284 of this Society's Journal 
for September 1951. The minutes also 
report the change of name of the Com- 
mission from International Commission on 
Illumination (I. C. I.) to Commission 
Internationale De L'Eclairage (C. I. E.) 
because of a conflict with the prior trade- 
mark of Imperial Chemical Industries in 
England. 

Decision was announced to hold the 
next meeting in 1955 in Switzerland. 
W. W. Lozier, National Carbon Co., Fos- 
toria, Ohio. 



A Television Policy for Education 

Carrol V. Newsom, Editor. Published 
(1952) by American Council on Education, 
1785 Massachusetts Ave., N.W., Wash- 
ington 6, D.C. i-xx + 267 pp. 6X9 
in. Price $3.50. 

Engineers, industrialists, administrators 
from government and education, lawyers, 
entertainers, producers and other leaders 
related their experiences and aspirations 
in developing, administering and program- 
ming educational television at the Tele- 
vision Programs Institute held at Penn- 
sylvania State College, April 20-24, 1952. 
Their papers have been abstracted in 
this well-edited volume, which includes a 
wealth of data on costs and practical steps 
to be taken. 

The Institute was sponsored by the 
American Council on Education which is 
a council of national educational associa- 
tions, universities, colleges, technological 
schools, state departments of education, 
and many other interested organizations 
including large public libraries. 

The Institute meeting was a serious task 



project with the following Advisory Com- t 
mittee : 

Milton S. Eisenhower, Chairman; Presi- 
dent, Pennsylvania State College. 
Arthur S. Adams, President, American i 

Council on Education 
Reverend Theodore M. Hesburgh, Presi- I 

dent, University of Notre Dame 
Armand L. Hunter, Director of Television j 

Development, Michigan State College < 
Francis Keppel, Dean, Harvard Graduate 

School of Education 

George E. Probst, Director of Radio, 
University of Chicago 
Mark C. Schinnerer, Superintendent of i 

Schools, Cleveland 
Ralph Steetle, Executive Director, Joint 

Committee on Educational Television \ 
The staff for the Institute was : 
Director: Carrol V. Newsom, Associate 

Commissioner for Higher Education, 

State of New York 
Assistant Director: E. Arthur Hungerford, 

General Precision Laboratory, Pleas- 

antville, N.Y. 
Consultant to the Director: Frederick W. 

Hoeing, New York 
Administrative Assistant: Mrs. Eunice Collins 

Parker, American Council on Educa- 
tion, Washington, D.C. 
From these and some 90 other par- 
ticipants there has been derived for this 
volume a substantial presentation of ad- 
dresses, panel discussions, reports and 
working papers. At the Institute there 
were 11 demonstrations which this book 
could only cite. 

Parts of the information here will not 
be new to some engineers and their com- 
munities but recent reports on experi- 
ence in New York state bear witness that 
a little more thorough dissemination may 
not be a dangerous thing. The State's 
Temporary Commission on the Use of 
Television for Educational Purposes re- 
cently delivered a report that the New 
York Times on March 1, 1953, called "an 
incredibly clumsy and shocking document." 
The Times thoroughly followed out its 
headline: "Incredible Document Re- 
port on Educational Video Is Shameful." 

The worth of a book like A Television 
Policy for Education may be further assessed 
by noting from the March 8, 1953, New 
York Herald-Tribune the concluding para- 
graph which columnist John Crosby wrote 



314 



[after supplying some of the known tele- falsifying every known fact about educa- 

vision cost figures : tional television." V.A. 

"The worst effects of the commission 

report are psychological New York State Audio _ V isual Communication 

has always been considered a leader in -r* 
affairs of this sort. Other legislatures and 

even private educators who had been This is a new quarterly issuing from the 

interested in educational TV might assume Department of Audio-Visual Instruction 

that New York has examined the thing (D AVI) of the National Education Associa- 

honestly and found it wanting. This is * 1 . 1 '***-- N ' W :' Washington 6 

untrue. Gov. Dewey has Lg been ^G. Articles dea mostly with film and 

television from a cultural point of view, and 

hostile to the Board of Regents and to a department of "Research Abstracts" 
everything they propose. The commis- prov ides precis of articles from other 
sion seems to have fallen all over itself to publications, dealing with films and tele- 
accommodate him, even to the extent of vision as training media. 

Journals Available and Wanted 



These notices are published as a service to expedite disposal and acquisition of out-of-print Journals. 
Please write direct to the persons and addresses listed. 

Available 

5 years (1947-51) in perfect condition plus the indexes for 1936-45 and 1946-50 and 
ncluding the 1949 High-Speed Photography, upon any reasonable offer to Vic Gret- 
zinger, 3547 Suter St., Oakland 19, Calif. 

1951-1952 Journals in excellent condition plus the Indexes for 1916-30, 1930-35, 1936-45 
and 1946-50; and including the 1949 High-Speed Photography. Write to K. G. Tsien, 
147-51 Charter Road, Jamaica 35, N.Y. 

Wanted 

Transactions 1, 6 and 7. Write Mrs. Dorothy Gelatt, Henry M. Lester, 101 Park Ave., 
New York 17, N.Y. 

April, 1934; May, 1936; Jan., 1938; Apr., 1939; June, 1940; July, 1941; Feb., 1944; 
Jan. and Aug. 1945; Jan., Feb. and Mar. 1946; Jan., 1947; Jan. and Feb. 1948; Mar. 
(in 2 parts), July, Aug. and Sept. 1949 and Feb. 1950. Write to Mr. Ishan Ghosh, Chief 
Technician, Kardar Productions 30, Government Gate Rd., Parel, Bombay 12, India. 

January and February 1946 Journals. Advise the Record Engineering Library, Radio 
Corporation of America, RCA Victor Division, 501 N. LaSalle St., Indianapolis, Ind. 



Sets of 74 standards (with index) in heavy three-post binders are available at $15.00 
(plus 3% sales tax on deliveries within New York City). Foreign postage is $0.50 extra. 
These sets include the latest standards approved: PH 22.1-1953, PH 22.84-1953, PH 
22.85-1953 and PH 22.92-1953. 

If all present owners of binders will communicate with the Society, these four new 
standards will be sent to them gratis. 

Single copies of any particular standard must be ordered from the American Standards 
Association, 70 East 45th St., New York 17, N.Y. Only complete sets are available from 
Society headquarters. 

315 



New Members 



The following members have been added to the Society's rolls since those last published. The 
designations of grades are the same as those used in the 1952 MEMBERSHIP DIRECTORY. 

Honorary (H) Fellow (F) Active (M) Associate (A) Student (S) 



Bird, Herbert Leslie, Chief Sound Engineer, 

UNESCO Mission, c/o Israel Motion Picture 

Studios, Ltd., Herzliya, Israel. (A) 
Brooks, Walter H., Editor, Managers' Round 

Table, Motion Picture Herald, 1270 Sixth 

Ave., New York, N.Y. (A) 
Byars, Taylor, Motion Picture Cameraman, 

Technical Director, Five Star Productions. 

Mail: 623 N. Lamer St., Burbank, Calif. 

(M) 

Harburger, Albert N., Film Editor and Pro- 
ducer, 630 Ninth Ave., New York 36, N.Y. 

(M) 
Hartzband, Morris, Free-lance Cameraman. 

Mail: 99-45 67 Rd., Forest Hills, N.Y. 

(A) 
Herman, Frank J., Processing Foreman, Geo. 

W. Colburn Laboratory, Inc. Mail: 7950 

Drexel Ave., Chicago, 111. (A) 
Killough, John G., Film Technician, Printer 

Foreman, Acme Film Laboratory. Mail: 

1444 Miller Dr., Los Angeles 46, Calif. (M) 
Kiteley, Raymond C., Motion Picture Camera- 
man, AF 11233560, Hq. Sq. Sec. Air Force 

Flight Test Center, Edwards Air Force Base, 

Calif. (A) 
Maguire, Frank J., Assistant Director, Medical 

Films, Audio Productions, 630 Ninth Ave., 

New York, N.Y. (A) 
Morse, Leonard P., Photographer, Visual Aids 

Technician, Pratt & Whitney Aircraft. 

Mail: 10 Dorothy St., Hartford 6, Conn. 

(A) 
Postal, Julius Bernard, Technical writer, Avion 

Instrument Corp. MaU: 435 Warwick St., 

Brooklyn 7, N.Y. (M) 
Robins, Ben W., Manager, Sound and Visual 

Design Group, RCA Victor Division. Mail: 

Haddon Hills Apt. 217-D, Haddonfield, N.J. 

(M) 



Schmit, Joseph W., Chemical Engineer, Techni- 
color Motion Picture Corp., 6311 Romaine, 
Hollywood 38, Calif. (A) 

Smith, Warren R., Motion Picture Technician, 
Warren R. Smith, Inc. Mail: 2560 Monroe- 
ville, Rd., Turtle Creek, Pa. (M) 

Sobin, Ben, Visual Education Service, Boy 
Scouts of America, 2 Park Ave., New York, 
N.Y._ (A)_ 

Valenzio, Victor E., Motion Picture Camera- 
man, U.S. Signal Corps Pictorial Center. 
Mail: 101-64 105 St., Ozone Park 16, N.Y. 
(A) 

Vanderhoek, E. F., Manager, Westrex Corp. 
(South-East Asia), 138 Robinson Rd., Singa- 
pore, Malaya. (M) 

Williams, J. Gordon, Transmission Engineer, 
Sound Services, Inc. Mail: 404 W. Valencia 
Ave., Burbank, Calif. (A) 

Zucker, Burton H., Assistant Cameraman, 
Camera Equipment Co., 1600 Broadway, 
New York, N.Y. (A) 

CHANGES IN GRADE 

Boden, William F., (S) to (A) 
Howard, William A., (A) to (M) 
Ushijima, Henry, (A) to (M) 

DECEASED 

Heacock, Frank C., Office Manager, Eastern 

Columbia. Mail: Box 136, San Gabriel, 

Calif. (A) 
Poulsen, Arnold, Director, Electrical Fono- 

Films Co., Meldahlsgade 5, Vesterport 1523, 

Copenhagen, Denmark. (A) 
Sherman, Harry, Labor Editor, International 

Projectionist Publishing Co., 19 W. 44 St., 

New York 18, N.Y. (M) 



Meetings 



Society of Motion Picture and Television Engineers, Central Section Meeting, Apr. 16, 

Encyclopaedia Britannica Film Studios, Chicago 

Symposium on Modern Network Synthesis, planned by Polytechnic Institute of Brooklyn, 
Apr. 16-18, Auditorium of Engineering Societies Bldg., New York 

International Symposium on Nonlinear Circuit Analysis, Apr. 23-24, information from 
Microwave Research Inst., 55 Johnson St., Brooklyn 1, N.Y. 

73d Semiannual Convention of the SMPTE, Apr. 27-May 1, Hotel Statler, Los Angeles 
National Association of Radio and Television Broadcasters, 7th Annual Conf., Apr. 28- 

May 1, Ambassador Hotel, Los Angeles 



316 



American Physical Society, Apr. 30-May 2, Washington, D.G. 
Acoustical Society of America, May 7-9, Hotel Warwick, Philadelphia, Pa. 
Society of Motion Picture and Television Engineers, Central Section Meeting, May 14, 

(tentative), Western Society of Engineers, Chicago 

Society of Motion Picture and Television Engineers, Southwest Subsection, May 20, 

Dallas, Tex. 

Society of Photographic Engineers, Third Annual Conference on Science in Photography 

and Photographic Instrumentation, May 20-22, U.S. Hotel Thayer, West Point, N.Y. 

Society of Motion Picture and Television Engineers, Central Section Meeting, June 11, 

Geo. W. Colburn Laboratory, Chicago 
American Physical Society, June 18-20, Rochester, N.Y. 

American Institute of Electrical Engineers, Summer General Meeting, June 29- July 3, 

Atlantic City, N.J. 

Biological Photographic Association, 23d Annual Meeting, Aug. 31 -Sept. 3, Hotel Statler, 

Los Angeles, Calif. 

The Royal Photographic Society's Centenary, International Conference on the Science 
and Applications of Photography, Sept. 19-25, London, England 

National Electronics Conference, 9th Annual Conference, Sept. 28-30, Hotel Sherman, 

Chicago 

74th Semiannual Convention of the SMPTE, Oct. 4-9, Hotel Statler, New York 
Audio Engineering Society, Fifth Annual Convention, Oct. 14-17, Hotel New Yorker, 

New York, N.Y. 

Theatre Equipment and Supply Manufacturers' Association Convention (in conjunction 
with Theatre Equipment Dealers' Association and Theatre Owners of America), 

Oct. 31 -Nov. 4, Conrad Hilton Hotel, Chicago, 111. 

Theatre Owners of America, Annual Convention and Trade Show, Nov. 1-5, Chicago, 111. 
National Electrical Manufacturers Association, Nov. 9-12, Haddon Hall Hotel, Atlantic 

City, N.J. 

75th Semiannual Convention of the SMPTE, May 3-7, 1954 (next year), Hotel Statler, 

Washington, D.C. 

I 76th Semiannual Convention of the SMPTE, Oct. 18-20, 1954 (next year), Ambassador 

Hotel, Los Angeles 



Employment Service 



Positions Available 

Wanted: Cameraman-Director, young 
man free to travel, to work in Southern 
states. Must be able to show sample 
footage. Send resume of experience and 
personal details in letter. Address replies 
to P.O. Box 1531, Louisville, Ky. 

Career Opportunity in Midwest: for 

competent photographer and cinematog- 
rapher experienced in 16mm production 
techniques to become permanently asso- 
ciated, as working partner, with well 
established, financially responsible com- 
pany producing medical and scientific 
films exclusively. Must be under 40, of 
excellent character, good personality and 
educational background. Write, stating 
age, education, experience, history of 
employment, marital status, salary ex- 
pected and where confidential corre- 



spondence can be addressed. Mervin 
W. LaRue, Inc., 159 E. Chicago Ave., 
Chicago 11, 111. 

Permanent Position in Southwest for 

experienced motion picture cameraman; 
must have sample interior and exterior 
footage to indicate ability. Write letter, 
giving re'sume of professional experience, 
to Susong Agency, 524 Commercial Bldg., 
Dallas, Tex. 

Position Wanted 

Resigned Feb. 1 as gen. mgr., charge of 
production, large southern film studio. 
15 yrs. experience as prod, mgr., editor and 
cameraman, 16mm and 35mm. Married, 
37, college grad. References and resume' 
on request Harlan H. Mendenhall, 5141 
Bataan St., Houston, Tex. 



317 



New Products 



Further information about these items can be obtained direct from the addresses given. 
As in the case of technical papers, the Society is not responsible for manufacturers' state- 
ments, and publication of these items does not constitute endorsement of the products. 



ROTATING MIRROR 



NTERNAL LENS 




This is a new research camera, Model 168, 
developed by Beckman and Whitley, Inc., 
San Carlos, Calif., and described here with 
data supplied by Jack Chambers, Director of 
Research of that organization. 

This smear or streak type of camera images 
the event to be studied through the vertical 
slit at the upper left, then the rotating stainless- 
steel mirror wipes the image onto film at a 
sweep rate of 0.327 to 5.466 mm//isec. Either 
4 X 5 in. or 4 X 10 in. film is used. Inside 
the upper right section of die housing below the 
vertically projecting knob is the mirror which 
by the knob is secured in a vacuum-tight round 
access panel. 

The schematic shows how a self-luminous 
image is projected through the slit by an external 
condenser lens which is not part of the instru- 
ment. The internal lens forms an inverted 
nonmagnified image of the slit in the film plane 
after reflection from the mirror surface. 

The slit width is 0.010 in. The slit plate is a 
piece of flat glass with an opaque coating on one 
side through which the slit line has been ruled. 
The shutter is a guillotine type to work at 
high speed to prevent multiple exposures, with 
exposure time of 1/200 sec. The mirror has a 
,3f X 4 in. reflecting surface. Temperature- 
monitoring thermocouples are mounted on the 
outer races of upper and lower mirror bearings. 

As shown in the schematic, synchronizing of 
the exposure with actuation of the test object is 
accomplished through the pip-generating com- 
mutator seen on top of the mirror. Each turn 
of the mirror produces a 15-v minimum pulse 



EXTERNAL 
CONDENSER - 
LENS 



TEST OBJECT- 



SWCE? BHUfllOH - HIRfcOR ROTiaiNG 




TIME RESOLVED 
STREAK 

FCTMmo 



which is used to trigger and synchronize the 
test object so that the image will be positioned on 
the film. 

A typical recording is depicted in the last 
illustration. At the extreme left is a slit image 
photographed for the purpose of determining 
the position of the test object, a gas-filled glow- 
discharge tube. To the right is a second slit 
image also made with the mirror stationary. 
This image constitutes the time base from 
which measurements are made to the streak 
image at the extreme right. This image is 
produced with the mirror spinning and a 
synchronized high-voltage pulse applied to the 
tube. 



318 




The Auto Camera Mark 3 has been 
announced as available from J. A. Maurer, 
Inc., Photographic Instrumentation Div., 
37-01 31st St., Long Island City 1, 
N.Y. Designed as a low-cost, lightweight, 
compact recording camera, it was origi- 
nally provided for aircraft use by D. Shack- 
man & Sons Ltd., London, England. It is 
8| in. long, 3f in. wide, 4 in. high and 
weighs 6 Ib. A spring motor drives the 
camera to transport and expose 21 ft of 
35 mm film, the cycle being initiated by 
either a 12- or 24-v d-c pulse. The film 
is held in special cassettes. Five shutter 
j speeds, from 1/10 to 1/200 sec, and 
| "time" exposure, are provided. The 
standard lens supplied is a 36-mm focal 
length, //3.5 in a graduated focusing 
mount. Other lenses are available and 
special models of the camera incorporate 
6- or 9-in. lenses. 

There are two models to provide alter- 
natives of 200 pictures 1 in. X 1 in. or 
300 pictures f in. X 1 in. Accessories 
permit time-lapse recording, photomicrog- 
raphy, normal and stereo photomacrog- 
raphy, aircraft instrument and chemical 
experiment recording. 



One-hundred and seventy-five tele- 
vision stations authorized to start opera- 
tions during 1953 are listed in the 16th 
edition of Television Factbook, published 
by Television Digest, Wyatt Bldg., Wash- 
ington, D.G. This figure covers all the 
post-freeze new television stations au- 
thorized through Jan. 3, 1953; of the 175, 
48 are VHF and 127 UHF. Television 
Factbook, a semiannual reference guide, 
has 268 pages in the new edition and costs 
$3.00. A new feature in this issue is a 



group of directories of community antenna 
systems, theater-television installations, and 
FCC channel allocation tables with priority 
lists. Also included are revised and 
brought-up-to-date directories of: engi- 
neers and attorneys; manufacturers of 
television equipment, including receivers, 
transmitters, tubes, theater-television, and 
community antennas; program sources; 
FCC personnel; station representatives; 
trade associations; unions; and publica- 
tions. 




A new 3-in. //3.5 telephoto lens for 16mm 
motion picture cameras has been an- 
nounced by Bell & Howell Co., 7100 
McGormick Rd., Chicago 45, 111. 

The new lens, which replaces the B & H 
3-in. //4 Telate lens, has all air-glass sur- 
faces coated. Accuracy of focusing is 
secured by matching the lens and focusing 
scale within 1 %. The lens carries an easy- 
to-read, standard spread-out iris scale with 
a range from //3.5 to //22. Click stops 
assure positive setting and prevent acci- 
dental changes of the diaphragm opening. 
The depth of field scale is filled in red for 
readier identification. Distances are cali- 
brated in feet from film. Supplied with 
the lens are a metal lens cap and a sun- 
shade. The sunshade serves also as a 
filter-holder. The retail price of the new 
lens, which is now available from Bell & 
Howell dealers, is $79.95. (Federal excise 
tax is not applicable.) 



319 



/ / 

.* 




Slide Rulc> Body 



This is the first working model of the 
Revised Slide Rule for complete analysis 
of high-speed motion picture data. 

In mechanics research, rapid methods for 
quantitative analysis of high-speed motion 
picture data are highly desirable. A first 
step in this direction was a slide rule 
developed at Springfield Armory in 1950, 
described before the Society at its 1951 
Spring Convention and in a paper pub- 
lished in the June 1951 Journal. That 
model was essentially planned for deter- 
mining the operation time and cyclic rate 
of moving parts, with additional use for 
some precalculation tasks. 

Recently, a more general slide rule has 
been developed. It and some of its 
possibilities were described in detail before 
the Society at its 1952 Fall Convention. 
It has been designed to allow for determi- 
nation of displacement, velocity and accele- 
ration of moving parts. It consists of 
slide rule body, three slides and an in- 
dicator, and carries scales for camera speed, 
number of frames, time, length of film 
interval, cyclic rate, object displacement, 
image displacement, velocity, acceleration 
and depth factor. The proper setting 



of the three slides is determined by the 
camera speed, the magnification and the 
length of film interval selected. The 
indicator is used for aligning corresponding 
values in various scales, such as number of 
frames vs. time, object displacement vs. 
image displacement, displacement vs. ve- 
locity, and velocity change vs. acceleration. 
In addition, a data sheet has been de- 
veloped to serve as a guide for the sequence 
of measuring and calculating operations 
and for plotting the results. 

The slide rule and the data sheet have 
been used at Springfield Armory over a 
period of some months and have proved 
successful, with much time saved in cal- 
culations. A computing aid, after brief 
training, can perform the whole operation, 
and it appears that errors in numerical 
calculations are reduced. The slide rule 
is expected to be of special value to small 
installations where expensive computing 
equipment is not justified. 

Manufacture and sale of the slide rule 
on a commercial basis have been post- 
poned until the pending patent application 
is cleared. Further information is avail- 
able from Springfield Armory, Atten: 
Karl W. Maier, Springfield 1, Mass. 



A new edition of the Society's Test Film Catalog is now available at no charge from the 
Society's headquarters. It covers 27 different test films, 16mm and 35mm, for use by 
theaters, service shops, factories and television stations. These test films have been 
developed by the SMPTE and the Motion Picture Research Council. 



320 



Recommendations of the National Television 
System Committee for a Color Television Signal 

By A. V. LOUGHREN 



I. INTRODUCTION 



The work of the National Television 
System Committee on color television 
has probably constituted a fundamental 
and major advance. This report of the 
NTSC's accomplishments is the author's 
view, and not an official statement on 
behalf of that Committee. 

To assist in the understanding of the 
significance of the several elements in 
the total contribution, a review of some 
of television's history is useful. 

In 1949 the Federal Communications 
Commission issued a Notice of Proposed 
Rule Making 1 dealing with color tele- 
vision, and held a hearing on the subject 
which lasted many months. The Com- 
mission appears to have had two pur- 
poses in this course of action. First, it 
certainly wanted to bring about a color- 
television service to the public at an 
early date. Second, and perhaps from 
its point of view, equally important, it 
wanted to establish whether or not color 
television would require different treat- 
ment in frequency allocation than did 
monochrome television. It had become 



Presented on October 6, 1952, at the So- 
ciety's Convention at Washington, D.G., 
by A. V. Loughren, Hazeltine Corp., Little 
Neck, L.I., N.Y. The author has reviewed 
the entire Convention transcript and 
brought salient points from the oral discus- 
sion into this presentation, brought up to 
date for the Journal. 



apparent before the issue of the notice 
that the 12 channels then available for 
television were insufficient to satisfy the 
Commission's objective of a nationwide 
and competitive television broadcast 
service. Additional channels in excess 
of 50 were going to be required. An un- 
committed portion of spectrum known 
to be reasonably suitable for television 
broadcasting lay below 1000 me, suf- 
ficient in extent for 70 added channels if 
6 me was sufficient channel width. No 
other frequency space gave promise of 
being available within the foreseeable 
future for television broadcasting. This 
state of affairs is shown in Fig. 1 . The 
Commission was faced, therefore, with 
the difficulty "If color television cannot 
be done in a 6-mc channel, we must 
either have less channels than needed 
for a nationwide competitive service or 
we must postpone color television into 
the indefinite future"; and there had 
been evidence, in 1947, that color-tele- 
vision broadcasting of quality compa- 
rable to monochrome television would 
require 12- to 15-mc channels. 

The Commission concluded, from the 
evidence presented in the hearing, that 
color television could be broadcast in a 
6-mc channel, that the field-sequential 
method of color-television broadcasting 
was the only method which at that time 
had been demonstrated both to give pic- 



April 1953 Journal of the SMPTE Vol. 60 



321 



VHP 

12 Channels 
of 6 Me 

in 

Commercial 
Use 



UHF 
70 Channels 

of 6 Me 

(or 35 of 12Mc) 

Known to Be 

Readily Usable 




Me* 54 88 174 216 470 890 



Remote UHP 

or Higher 
Usable for TV 
Broadcasting Only 
after Long Development 
If Ever 



2500 



Fig. 1. Existing and potential television allocations, as of 1949; 
immediate need, 70 or more channels. 




Panels 11 & 11-A 
Subjective Aspects 



Color Transcription 



ORGANIZATION OF NTSC 

Fig. 2. Organization of National Television System Committee. 



tures with some prospect of viewer ac- 
ceptance and to be practicable with ap- 
paratus then readily available, and the 
Commission adopted that method. 2 - 3 

In response to the Commission's 
notice of the hearing a large number of 
interested parties in the television broad- 
casting and television apparatus manu- 
facturing industries established a Na- 



tional Television System Committee to 
provide a means for the industry to assist 
the Federal Communications Commis- 
sion by making information available 
to it in the course of the hearing. Be- 
fore, during, and following the hearing, 
some of the members of this committee 
had made substantial progress with both 
the theoretical and practical aspects of 



322 



April 1953 Journal of the SMPTE Vol. 60 



a method of color-television broadcast- 
ing quite different from the field-sequen- 
tial method. In the fall of 1950 the 
NTSC set up a six-member Ad Hoc 
Committee to investigate the progress of 
these developments and to report back a 
recommendation as to whether or not 
NTSC should work actively toward the 
completion and experimental verifica- 
tion of a color-television signal based on 
all of these developments. In the spring 
of 1951 the Ad Hoc Committee turned 
in its report. The committee found 
that the developments had indeed 
reached the point where it seemed highly 
likely that further work would lead to 
a most advantageous form of color-tele- 
vision broadcast signal. The commit- 
tee's report recommended that the 
NTSC proceed actively to study the sig- 
nal, to arrange for such tests as were nec- 
essary to select the signal characteristics, 
and to pursue a program of field tests of 
such scope as to show clearly beyond any 
doubt whether or not the signal result- 



ing from these choices was in fact pos- 
sessed of the advantages claimed for it. 
The National Television System Com- 
mittee approved the report of its Ad 
Hoc Committee and it proceeded in ac- 
cordance with that report to establish 
additional panels organized specifically 
for the several tasks involved, as pre- 
sented in Fig. 2. Panels 13 and 14 are 
responsible for establishing the signal 
specifications, Panels 15, 16 and 17, for 
the independent testing of the recom- 
mendations of the first-named two pan- 
els, and the remaining panels provide 
for important functions auxiliary to those 
already described. In the ensuing 
months the signal specification has been 
established, been subjected to extensive 
studies in both laboratory and field, 
and has been modified to make certain 
minor improvements. The committee's 
work is now at the point where the final 
field tests are expected to start in the 
immediate future. 



II. THE BASIC PRINCIPLES 



Out of the work in this field certain 
principles have emerged which appear to 
be basic to a system of color-television 
broadcasting if that system is to show 
full utilization of the portion of the fre- 
quency spectrum allocated to it. Among 
these principles are the following: 

1. The electrical quantities in the sig- 
nal shall represent the transmitted color 
in terms of its luminance, dominant 
wavelength and purity; alternatively, 
these may be thought of as brightness, 
hue and saturation. 

2. The spectrum space occupied by 
the luminance information must be 
shared by the color information without 
adversely affecting the transmission of 
either. 

3. The color-television signal must 
provide satisfactory black-and-white pic- 
tures on the present monochrome sets if 



color broadcasting is to grow rapidly 
into an important public service. 

These principles are worth examina- 
tion in some detail. 

The Electrical Description of Color 

There are many ways of including in 
an electrical signal three quantities 
which when taken together can be used 
to represent a color. Only two of these 
need concern us here. In the first of 
these, separate elements of the signal are 
used to represent the respective intensi- 
ties of three primary colors for ex- 
ample, red, green and blue -- which 
taken together will produce a resultant 
color which to the eye is a match for the 
original color. This method of electri- 
cally describing a color appears to be 
simple and straightforward, and it can 
be shown to have the further advantage 
that the three indications may be pre- 



A. V. Loughren: Color Television 



323 




O.I 



0.25 



0.5 1.0 2.0 

CROSSOVER FREQUENCY IN MEGACYCLES 



fSKILLED a 
J INSTRUCTED 
lOBSERVERS 

^INSTRUCTED 
GROUP 



Fig. 3. Comparison of "Satisfactory Picture" curves for various skills. 
Courtesy, Proc. I.R.E. (Ref. 6, Fig. 7). 



sented successively rather than simul- 
taneously without altering too greatly 
the visual result, at least over a limited 
range of conditions. It is therefore pos- 
sible by fairly easy modifications of a 
black-and-white television system to ar- 
range for the display of one field of the 
picture in each of the primary colors suc- 
cessively. Theoretical and experimen- 
tal studies of such a system appear to 
have established the following system 
properties : 

1. If the color-television system is to 
exhibit essentially the same resolution 
as that of a typical monochrome system, 
it must present as much information for 
each of the three primary colors as the 
monochrome system presents; it must 
therefore transmit three times as much 
total information and use three times as 
much bandwidth. 

2. If the color system confines itself 
to the same bandwidth as the mono- 
chrome system, it must as a practical 
matter exhibit decreased resolution and 
must also operate with different scan- 



ning standards than the monochrome 
system. 

Considering now the electrical de- 
scription of color in terms of brightness, 
hue and saturation, we note first that 
these quantities must be presented simul- 
taneously; there is no mechanism in the 
eye which permits these quantities to be 
accepted successively in the fashion that 
persistence of vision permits red, green 
and blue intensities to be accepted suc- 
cessively. Next we find that the visual 
acuity for the brightness component of 
color, that is, the ability to perceive fine 
detail, is exactly the same as that for the 
brightness component of a monochrome 
image, but visual acuity for changes of 
color unaccompanied by changes of 
brightness is very much poorer than the 
eye's acuity for brightness detail; this 
indicates that the amount of information 
which must be transmitted to add color 
to a black-and-white picture is very 
much smaller than the amount of infor- 
mation required to produce a satisfac- 
tory black-and-white picture in the first 
place, if the color is added by informa- 



324 



April 1953 Journal of the SMPTE Vol.60 



tion which describes hue and satura- 
tion. 4 - 5 

Panel 11 of the NTSG conducted a 
careful set of experiments, 6 the results of 
which are summarized in Fig. 3. In 
making these tests all pictures were pre- 
sented with a bandwidth for the lumi- 
nance component extending out to 4 me. 
The additional information required to 
insert color into the picture may be 
thought of as "color-difference" sig- 
nals; two color-difference signals taken 
together can be used to tell how far 
from gray and in what direction from 
gray a particular color is, relative to a 
gray of exactly the same luminance. 
If the color-difference signals have a 
narrower bandwidth than the luminance 
signal, then color gradations are not as 
sharp on the picture although the full 
sharpness of luminance gradations is pre- 
served. The horizontal coordinate (on 
the chart, called "crossover frequency," 
in reference to the specific apparatus of 
the test) is the amount of bandwidth 
devoted to each of the two color-differ- 
ence signals in a particular observation. 
The vertical coordinate refers to the per- 
centage of the number of observations 
in which the various classes of observers 
rated the picture quality as "satisfac- 
tory." The curves indicate that even 
with a luminance bandwidth of 4 me, 
and the observers permitted to sit where 
they wished, there was no significant 
improvement in pictorial quality ob- 
tained by increasing the bandwidth de- 
voted to color-difference signals above 1 
me. The curves also show that this 
statement is true both for skilled and for 
lay observers. 

In the tests made by Panel 11 both 
color-difference signals had the same 
magnitude in any one observation. Sub- 
sequent tests, the results of which have 
not yet been published, have indicated 
that along a particular direction of 
color difference the bandwidth may be 
decreased to approximately ^ me without 
noticeable impairment of the picture, 
especially if the bandwidth in another 



direction is retained at the value of 1 me 
or perhaps slightly greater. 

We also note that the eye responds 
only slowly to changes in color; thus 
small rapid fluctuations about a correct 
color value go unperceived, whereas 
corresponding fluctuations about the 
correct brightness value would be per- 
ceived immediately as flicker. This 
latter point indicates that greater ex- 
posure to electrical disturbance may be 
permitted the color component of the 
signal without impairment of perform- 
ance than may safely be permitted to 
the monochrome component of the sig- 
nal. 

Sharing of the Channel 

The luminance component of a color- 
television signal has the task of control- 
ing, element by element, the brightness 
of the received image. It is most impor- 
tant for successful color-television trans- 
mission that this task be well discharged. 
The standards of the Federal Communi- 
cations Commission for monochrome 
television provide the means for dis- 
charging this task well, and the use of 
these standards should certainly be ex- 
amined as a basis for a successful color- 
television system. 

It has been known for a long time that 
a normal monochrome television signal 
by no means completely fills its chan- 
nel; 7 - 8 the spectrum consists in general 
of a component at each harmonic of the 
line-scanning frequency, with each such 
component being accompanied by a 
cluster of smaller components spaced 
from the main component by the field- 
scanning frequency. Figure 4 illus- 
trates a small section of the spectrum, 
showing the region in the vicinity of the 
eighty-second and eighty-third harmon- 
ics of the line-scanning frequency. The 
dotted line half-way between the groups 
illustrates the absence of any signal in- 
formation at this region. The signal 
spectrum consists therefore of groups of 
components at the successive harmonics 
of the scanning-line frequency, or bet- 



A. V. Loughren: Color Television 



325 




FREQUENCY 



LINE X 



1/2 LINE X 



82 



164 



165 



83 



166 



Fig. 4. Small section of the spectrum, region in the vicinity of the 82nd and 83rd 
harmonics of the line scanning frequency. 

Courtesy, Electronics (Ref. 9, Fig. 1). 



ter, at the even harmonics of one-half 
the scanning-line frequency, and gaps 
corresponding to the odd harmonics of 
one-half the scanning-line frequency. 
We see, therefore, that a second television 
picture could be transmitted within the 
same spectrum occupied by our lumi- 
nance picture if its signal components 
could be so transformed as to lie at odd 
harmonics of the scanning-line frequency. 
It is also found that even if compo- 
nents are artificially injected into the 
signal spectrum in positions correspond- 
ing to the dotted line of Fig. 4, a normal 
television receiver will not reproduce 
these components. The diagram of Fig. 
5 illustrates the reasons for this. In the 
upper line of the figure a signal is shown 
at the third harmonic of the line-scanning 
frequency and therefore at the sixth, an 
even harmonic of half the line frequency. 
Three full cycles of this signal appear in 
traversing a single scanning line, and the 
next scanning line shows three full cycles 
again. The section at the right of the 
figure is labeled "Line 3" or "526," and 
of course line 526 is the first line of the 
next picture, and therefore lies exactly 
on top of line 1 one-thirtieth of a second 
later. The second horizontal line in the 
diagram shows how line 1 and line 526 



add directly to produce an augmented 
result. In the lower half of the diagram 
a signal is shown at the third harmonic 
of one-half the line frequency. This 
odd harmonic signal repeats in line 3 
and also in line 526 exactly opposite in 
polarity to the position which it took in 
line 1. The lowest section of the dia- 
gram at the right shows that when line 1 
and line 526 are superposed, the brighter 
than average regions in line 1 land on 
darker than average regions in line 526, 
and vice versa, and therefore the total 
contribution to brightness tends to be 
the same all along the line ; the compo- 
nent is therefore not effectively repro- 
duced. If the reproducing device is 
linear, and if the eye remembers fully 
for two picture intervals, then this can- 
cellation is exact and complete; in the 
practical case neither of these require- 
ments is fully met, and so the cancella- 
tion is theoretically incomplete but of 
great practical value. 9 

It is worth emphasizing that we have 
thus established the fact that a good 
monochrome television signal, suitable 
for use additionally as the luminance 
component of a color signal, has gaps 
in it in which additional information 
may be transmitted without affecting 



326 



April 1953 Journal of the SMPTE Vol. 60 



MODULATION 
FREQUENCY 
IS EQUAL TO 



A- EVEN HARMONIC 
OF \fl [LINE FREQ] 



B-ODD HARMONIC 
OF 1/2 [LINE FREQ.] 



LIME 1 



LINE 1 PIUS LINE 526 
ADO IN PHASE 



LINE 1- 

PLUS 

LINE 526 

CANCEL 

TIME * 

Fig. 5. Principle of interference cancellation by frequency interleaving. 

Courtesy, Electronics (Ref. 9, Fig. 2). 



LINE 3 OR 526 



t 
\ ,' \ I \ / 

N..V X..X v..X 



the use of that signal as a luminance sig- 
nal either for color or for monochrome 
receivers. 

Compatibility Considerations 

We have seen that the signal for mono- 
chrome television and the luminance 
component of a color-television signal 
must satisfy the same set of require- 
ments ; we have also seen that idle inter- 
vals exist in the transmitted band of this 
signal in which another signal may be 
placed, and we shall presently find that 
the color components of the color-tele- 
vision signal may be placed there; we 
have also seen that a signal placed in 
these idle intervals is essentially invisible 
on a black-and-white receiver. There 
is no disadvantage then in adopting for 
the luminance component of a color- 
television transmission identically the 
same standards which have been found 
suitable for a high-grade monochrome 
transmission. And we should bear in 
mind that the performance capabilities of 
the black-and-white television signal 



represented by the FCC's standards, are 
likely to equal or exceed any demand for 
performance to be encountered in the 
foreseeable future. 

But there is more to this than the mere 
absence of a disadvantage. If our color- 
television signal consists of a luminance 
component conforming exactly to the 
present monochrome standards and an 
interleaved color component which is 
essentially invisible on monochrome re- 
ceivers, then our color-television broad- 
cast is also an acceptable black-and- 
white broadcast, and the potential audi- 
ence for any such broadcast includes not 
merely those who have at the time of the 
broadcast equipped themselves with 
color-television receivers, but also all 
viewers equipped with black-and-white 
receivers. A signal for color television 
which bears this relation to an existing 
signal for monochrome television, is said 
to be compatible with the monochrome 
signal; the importance of compatibility 
in bringing about early and widespread 
adoption of color television, once the 



A. V. Loughren: Color Television 



327 



color standards have been accepted and 
established by the Federal Communica- 
tions Commission, is thus seen to be ex- 
tremely great. With compatibility, color 
television can be given a flying start by 
its parent, monochrome television ; with- 
out compatibility, color television must 
start from a complete standstill; it must 
face the terrible economic difficulty rep- 
resented by the situation "too few 
viewers, therefore very little sponsor 
money, therefore poor programs, there- 
fore too few viewers." 

Important though compatibility is to 
the rapid introduction of color tele- 



vision, it is perhaps not important enough 
to justify of itself the adoption of a color- 
television system markedly inferior to a 
system which could be developed dis- 
regarding compatibility. We should 
note however that we are not faced with 
this question; instead, we have the situ- 
tion in which the basic technical consid- 
erations important in the design of a 
color-television signal coincide with 
those controlling the design of a mono- 
chrome signal, and in satisfying these 
considerations we find compatibility as 
an automatic by-product. 



III. GENERATION OF THE SIGNAL 



A block diagram of the essential ele- 
ments of the transmitting apparatus is 
shown in Fig. 6. A camera at the left 
generates three signals which collec- 
tively represent the luminance, hue and 
saturation of the scene, element by ele- 
ment; in the example shown the actual 
camera outputs correspond to the red, 
green and blue components. Gamma 
correction to improve the signal-to- 
noise ratio and incidentally to match ap- 
proximately the receiver picture-tube 
curvature may be introduced at this 
point. 

The resulting signals are electrically 
added, in the proportions of their re- 
spective luminances, to form the mono- 
chrome or luminance signal. Normal 
adjustment for the system is such as to 
make the three signals from the camera 
equal if the color represented is that cor- 
responding to standard Illuminant C; 
under this condition the luminance sig- 
nal E Y ' is also equal to the gamma cor- 
rected voltages E G ', E R ' and E B '.* 

The negative of the luminance signal 
is developed by the phase inverter; in 
the two adders shown below it, this sig- 
nal is added to the red and the blue color 



* See the Appendix for definitions of sym- 
bols and formulas not explained in the 
text and illustrations. 



signals, respectively, thus forming in the 
adder the two color-difference signals 
corresponding to these two primaries. 
Examining the red color-difference signal 
in some detail, it is noted that this signal 
was generated by addition and subtrac- 
tion of the signals originally produced by 
the camera. These signals, of course, 
have their components at even harmonics 
of one-half the line frequency like the 
components shown solid in Fig. 4; of 
course, addition and subtraction of such 
components does not change their fre- 
quency into the positions of the low- 
visibility, odd-harmonic components, 
and yet we want these components to 
be translated into odd-harmonic com- 
ponents so that they may be inter- 
leaved between the components of the 
luminance signal for transmission. If 
we generate a subcarrier frequency at 
such a frequency value as, for example, 
the 455th harmonic of one-half the line 
frequency, it will be a low-visibility sig- 
nal, like the dotted line in Fig. 4. If we 
then use our red color-difference signal 
to modulate this subcarrier, the side- 
bands which appear with the subcarrier 
as a consequence of the modulation proc- 
ess, will be separated from the subcarrier 
frequency by the same interval that 
separates the original sidebands from 
zero frequency; in other words, all of 



328 



April 1953 Journal of the SMPTE Vol. 60 



MONOCHROME 

SIGNAL 
GENERATOR 



0-4.0 me 




NOTE 

VALUES OF COLOR BANDWIDTHS 
ARE THE MINIMUM SPECIFIED. 



sin 0)t 
COLOR SUBCARRIER GENERATOR 



| 

Fig. 6. Block diagram of the essential elements of the transmitting apparatus. 

Courtesy, Electronics (Ref. 9, Fig. 5). 



K- ft, .i.f H 

I ^^ 




A- MONOCHROME ONLY, 
"f m" (0*4 me). 



B- COLOR ONLY "fc" (O'l.Smc) AND 
COLOR SUB-CARRIER FREQUENCY, 
M f,c" (3.5 4- me). 



C-COLOR AFTER MODULATION 
"f, e f c ." 



D-MONOCHROME "fm"(0-4mc) 
ft MODULATED COLOR M f $c f c " 
AFTER BANDWIDTH LIMITING 
TO (2-4 me). 

ft" LINE FREQUENCY* I5.7500J 

fn, AND f c ARE EVEN HARMONICS OF ONE-HALF LINE-FREQUENCY Ci 2n^l 

f sc IS ODD HARMONIC OF ONE -HALF LINE-FREQUENCY Qt 455 f 3.583125 me'.]. 

[jtcfc3 IS ODD HARMONIC OF ONE-HALF LINE-FREQUENCY [it [20 - 71 y~j. 

n,m,k < ORDER OF HARMONIC = 1,2,3,4,5 ETC. 

Fig. 7. Band-sharing by monochrome and color signals. 



the components of the red color-differ- 
ence signal which were generated as 
even harmonics of one-half the line fre- 
quency, will appear as sidebands of the 
subcarrier at odd harmonics of one-half 
the line frequency, and will therefore 



exhibit low visibility if they are applied 
to any normal television reproducing 
system. This frequency transformation 
is illustrated in Fig. 1 '. Row A shows 
the luminance signal and its components 
in solid lines to represent their being 



A. V. Loughren: Color Television 



329 



even harmonics of one-half the line fre- 
quency. Row B shows a color-differ- 
ence signal as generated, again with 
solidly drawn components representing 
even harmonics of one-half the line fre- 
quency. Row C shows this same signal 
appearing as modulation on a subcarrier 
f te . The spacings of the components are 
still the same but the subcarrier itself 
being an odd harmonic, all of the other 
components must themselves be odd 
harmonics also. Finally, Row D of the 
drawings shows the solid lines of the 
first section and the dotted lines of the 
third section interleaved as they are in a 
normal transmission. 

Some other characteristics of the sig- 
nal may be noted by further reference to 
Figure 6. For example, since the red 
signal and the signal E Y ' are equal on 
white, the output of the adder in the red 
channel, the red color-difference signal, 
must equal zero on white. It will have 
an amplitude which we may call posi- 
tive when the scene is red and an ampli- 
tude which we may call negative when 
the scene is the complement of red, or 
cyan. Now, if the modulator of Fig. 6 
is a balanced modulator, it can be ar- 
ranged so that its output will also vanish 
on white; it is advantageous to design 
the system in this fashion, since this re- 
duces still further the residual percepti- 
bility of the low-visibility components 
represented by the dotted lines of Figs. 
4 and 7, especially in picture highlights 
which are generally white or relatively 
unsaturated color. We find, however, 
that a transmission whose output van- 
ishes when the modulating signal van- 
ishes, is a transmission in which the car- 
rier frequency has been suppressed, 
leaving only the modulation sidebands. 
For correct detection of such a transmis- 
sion the carrier must be resupplied at the 
detector of the receiver, and we shall 
provide a synchronizing pulse in the 
transmission to enable receivers to gener- 
ate a suitable carrier. The resupplying 
of a properly synchronized carrier in the 
receiver offers us another advantage: 



it permits us to distinguish successfully 
between phase modulation and ampli- 
tude modulation of a single subcarrier; 
alternatively, we may say that it permits 
us to modulate two subcarriers at the 
same frequency 90 apart in phase and 
transmit their sidebands over the same 
circuit and yet distinguish each set of 
sidebands from the other in the receiver. 
It will be seen that we make use of ex- 
actly this property to permit the trans- 
mission of the blue color-difference signal 
as modulation sidebands on the very same 
subcarrier frequency as the red color- 
difference signal, with the two subcar- 
riers differing merely by 90 in phase at 
the subcarrier frequency. 

Finally, the two sets of subcarrier side- 
bands representing respectively red color- 
difference information and blue color- 
difference information, are combined 
and are then added to the luminance 
signal as illustrated frequency-wise in 
Fig. 7. The resulting signal is then ap- 
plied to the transmitter. Let us note 
again that this signal is in all respects a 
perfectly normal monochrome television 
signal to which there has been added, in 
a fashion which makes it essentially in- 
visible on normal monochrome receiv- 
ers, the color-difference information 
which can be used in a color television 
receiver to reconstruct the original image 
in full color. 

The transmission of two independent 
sets of modulation sidebands based on 
the same subcarrier frequency over the 
same circuit requires that the transmis- 
sion be on a double sideband basis if the 
two sets of modulation components are to 
be separated at the receiver. Conse- 
quently, the maximum frequency which 
we may choose for the subcarrier is a 
frequency enough lower than the top fre- 
quency of the expected passband of the 
system to satisfy this requirement for 
double sideband transmission. Now, if 
one of the modulation components has a 
bandwidth of a half megacycle, while 
the other had a bandwidth of a mega- 
cycle or greater, the necessity for double 



330 



April 1953 Journal of the SMPTE Vol. 60 



sideband transmission extends out only a 
half megacycle away from the nominal 
frequency of the subcarrier; this fact, 
together with the practical requirements 
of receiver construction, dictates a sub- 
carrier frequency in the vicinity of 3.5 
me. The exact value of the subcarrier 
frequency must be, of course, an odd 
harmonic of one-half the line-scanning 
frequency; there is an additional minor 
advantage to be gained by making the 
difference between the subcarrier fre- 
quency and the frequency at which 



sound is transmitted be also an odd mul- 
tiple of one-half the line-scanning fre- 
quency. The choice of a subcarrier 
frequency of 3.579545 me satisfies these 
requirements without involving any 
change in the separation between picture 
and sound carriers as currently specified ; 
a decrease in line-scanning frequency of 
0.1% from the presently specified value 
results from this choice but this decrease 
is negligible compared to the tolerance 
presently permitted to the line-scanning 
frequency itself. 



*-COLOR SIGNAL 




2 3 4 MC 



COLOR SIGNAl* 



01 2 3 4 MC 
COLOR CHANNEL 



I 2 3 4 MC 
MONOCHROME CHANNEL E V 

t 




CHROMATICITY 
ONLY 



Fig. 8. Elements of one form of receiver for the proposed NTSC signal. 

Courtesy, Electronics (Ref. 9, Fig. 7). 



IV. THE COLOR RECEIVER 



Color-television receivers are not the 
direct concern of the NTSC; the Com- 
mittee's primary objective is the specify- 
ing of a signal for color-television broad- 
casting which is believed to be a sound 
basis for the founding of a nationwide 
color- television service, and the recom- 
mending of that signal to the Federal 
Communications Commission after ade- 
quate testing. (Receivers are of interest 
to the Committee indirectly, however, 
for two important reasons: first, it is 
essential that the color-television signal 



which the Committee recommends be 
suitable for use with receivers which can 
be built practically and sold commer- 
cially at prices which will interest the 
public; second, the verifying of the suit- 
ability of the Committee's proposed 
signal is an experimental process and 
must use both physical transmitters and 
physical receivers for that verification. 
A detailed discussion of receivers would 
therefore be inappropriate in this report 
of the work of the NTSC. A few words 



A. V. Loughren: Color Television 



331 



about the general scheme of receivers 
for use with the NTSG signal may, how- 
ever, be appropriate. 

Figure 8 shows the elements of one 
form of receiver for the proposed NTSC 
signal. In this receiver a three-gun 
picture tube has been used; other forms 
of display device are entirely possible. 10 
The receiver includes all of the usual 
elements of a monochrome television 
receiver such as radio and intermediate 
frequency circuits, detector, video fre- 
quency circuits to feed the picture tube 
grid, and the usual scanning and high 
voltage supply circuits for the picture 
tube. The color circuits are energized 
from a tap off the main video circuit of 
the receiver through a filter which passes 
only the portion of the video spectrum 
containing the color components. This 
selected signal includes the modulation 
sidebands which were generated in the 
transmitter by application of the original 
color difference signals as modulation 
to the color subcarrier; it also includes 
the components of the luminance signal 
which appear in the selected frequency 
band. The luminance components are, 
of course, even harmonics of half the line 
frequency while the color sidebands are 
odd harmonics. A local oscillator in the 
receiver reproduces the subcarrier fre- 
quency which was suppressed at the 
transmitter; it is kept accurately in 
phase by reference to the periodic "burst" 
of the color subcarrier frequency which 



is transmitted during the synchronizing 
interval. The local oscillator signal and 
the color signal selected from the re- 
ceiver output are both applied to the 
red color-difference demodulator; its 
output contains the beat between these, 
and jt is found that one component of 
that complex beat is the original video- 
frequency, color-difference signal. The 
frequency conversion occurring in the 
production of this beat signal has trans- 
formed all of the color modulation side- 
bands which were odd harmonics of one- 
half the line frequency as transmitted 
into even harmonics as the signal appears 
in the output of the demodulator. These 
even harmonics are, of course, suitable 
for producing a visible image on the 
cathode-ray tube and we therefore apply 
them to the red gun of that tube. Simi- 
lar procedure is followed with respect to 
the blue color-difference signal and the 
blue gun, the only difference being that 
the phase of the locally generated sub- 
carrier as applied to the blue demodula- 
tor is 90 away from the phase in which 
that signal is applied to the red demodu- 
lator. The green color-difference sig- 
nal is obtained by a proper addition of 
the red and blue signals, giving due re- 
gard to the algebraic signs of the signals 
and the required output; alternatively, 
it may be obtained by the use of a third 
demodulator if the phase of the local sub- 
carrier applied to that demodulator is 
properly chosen. 



V. CONCLUSION 



In its work on color television the Na- 
tional Television System Committee has 
now formulated a proposal for a color- 
television broadcasting signal.* The 
design of the signal is based upon careful 
study of the information need of the 
human viewer, and it is believed that the 
signal is capable of adequately satisfying 
that need. The signal is transmitted 



* See the Appendix. 



in the same 6-mc channel as our present 
signals and has the incidental but im- 
portant feature of being compatible 
with present monochrome television 
signals. Transmitting, networking, and 
receiving apparatus suitable for use with 
the signal is now becoming available in 
sufficient quantity to permit thorough 
field testing of the proposal. It is my 
expectation that the field test will show 
conclusively the suitability of the signal 
for color-television broadcasting. 



332 



AprU 1953 Journal of the SMPTE Vol. 60 



VI.jREFERENCES 



FCC Public Notice 49-948 of July 11, 
1949. 

FCC Public Notices No. 50-1064 and 
50-1065 of Sept. 1, 1950; FCC Public 
Notice No. 50-1225 of Oct. 11, 1950. 
P. G. Goldmark, J. W. Christensen 
and J. J. Reeves, "Color television 
U.S.A. Standard," Jour. SMPTE, 57: 
336-381, Oct. 1951. 
A. V. Loughren and C. J. Hirsch, 
"Comparative analysis of color TV 
systems," Electronics., 24: 92-96, Feb. 
1951. 

A. V. Bedford, "Mixed highs in color 
television," Proc. I.R.E., 38: 1003- 
1009, Sept. 1950. 

K. Mcllwain, "Requisite color band- 
width for simultaneous color-television 
systems," Proc. I.R.E. 40: 909-912, 
Aug. 1952. 

7. F. Gray, U.S. Patent 1,769,920; filed 
Apr. 30, 1929; issued July 8, 1930. 

8. P. Mertz and F. Gray, "A theory of 
scanning and its relation to the char- 
acteristics of the transmitted signal in 
telephotography and television," Bell 
Sys. Tech. J., 13: 464-515, July 1934. 
C. J. Hirsch, W, F. Bailey and B. D. 
Loughlin, "Principles of NTSC com- 
patible color television," Electronics, 25: 
88-95; Feb. 1952. 



6. 



9. 



10. C. S. Szegho, "Color cathode-ray tube 

with three phosphor bands," Jour. 

SMPTE, 55: 367-376, Oct. 1950. 
E. W. Herold, et al ; "Direct-view color 

kinescopes : a series of eleven papers," 

Proc. I.R.E., 39: 1177-1263, Oct. 

1951. 
D. G. Fink, "Phosphor-strip tricolor 

tubes," Electronics, 24: 89-91; Dec. 

1951. 
L. C. Jesty, British Patent 443,896; 

application date Oct. 6, 1934; issue 

date Mar. 10, 1936. 

Additional References 

D. G. Fink, "Alternative approaches to 
color television," Proc. I.R.E. 39: 1124- 
1134, Oct. 1951. 

W. T. Wintringham, "Color television and 
colorimetry," Proc. I.R.E. 39: 1135- 
1172, Oct. 1951. 

B. D. Loughlin, "Recent improvements in 
band-shared simultaneous color-tele- 
vision systems," Proc. I.R.E., 39: 1264- 
1279, Oct. 1951. 

N. Marchand, H. R. Holloway and M. 
Leifer, "Analysis of dot-sequential color 
television," Proc. I.R.E., 39: 1280-1287, 
Oct. 1951. 

R. B. Dome, "Spectrum utilization in color 
television," Proc. I.R.E., 39: 1323-1331, 
Oct. 1951. 



APPENDIX: Revised Specifications for Field Test of NTSC Compatible 
Color Television 



Test Specifications Group I 

1. The image is scanned at uniform 
velocities from left to right and from top 
to bottom with 525 lines/frame and 
nominally 60 fields/sec, interlaced 2-to-l. 

2. The aspect ratio of the image is 4 
units horizontally and 3 units vertically. 

3. The blanking level is fixed at 75% 
(2.5%) of the peak amplitude of the 
carrier envelope. The maximum white 



NOTE: These revised specifications were 
formally released on Feb. 4, 1953, by W. R. 
G. Baker, Chairman of NTSC, c/o General 
Electric Co., Electronics Park, Syracuse, 
N.Y., with the advice that a comprehensive 
field test would soon be inaugurated. Dr. 
Baker will welcome comments. 



(luminance) level is not more than 15% 
nor less than 10% of the peak carrier 
amplitude. 

4. The horizontal and vertical syn- 
chronizing pulses are those specified in 
Sec. 3.682 of Subpart E of Part 3 of the 
FCC Rules Governing Radio Broadcast 
Services (as amended April 11, 1952; 
effective June 2, 1952), modified to pro- 
vide the color synchronizing signal de- 
scribed in Specif. 21 (Group II of these 
specifications). 

5. An increase in initial light intensity 
corresponds to a decrease in the ampli- 
tude of the carrier envelope (negative 
modulation). 

6. The television channel occupies a 
total width of 6 me. Vestigial-sideband 



A. V. Loughren: Color Television 



333 



amplitude-modulation transmission is 
used for the picture signal in accordance 
with the FCC Rules cited in Specif. 4, 
above. 

7. The sound transmission is by fre- 
quency modulation, with maximum de- 
viation 25 kc, and with pre-emphasis 
in accordance with a 75-jusec time con- 
stant. The frequency of the unmodu- 
lated sound carrier is 4.5 me 1000 
cycles above the frequency of the main 
picture carrier actually in use at the 
transmitter. 

8. The radiated signals are horizon- 
tally polarized. 

9. The power of the aural-signal trans- 
mitter is not less than 50% nor more than 
70% of the peak power of the visual- 
signal transmitter. 

Test Specifications Group n 

10. The color picture signal has the 
following composition: 

E m = E Y ' + {^Q'sin (at + 33) 

+ /'cos(orf + 33)} 

where 

EQ' = 0.41 (E B ' - E Y ') * 

+ 0.48 (E R r - E Y '} 

Ei' = -0.27 (E B ' - E Y ') 

+ 0.74 (E* 1 - EY') 

E Y ' = 0.30 E R ' -f 0.59 E G ' 

+ 0.11 E tt ' 

The phase of the color burst is sin (to/ + 
180) 

Notes: For color-difference frequencies 
below 500 kc, the signal can be repre- 
sented by 



E. . E 



sin o>t + (E R ' 



cos at 



\ f 



In these expressions the symbols have 
the following significance: 

E m is the total video voltage, corre- 
sponding to the scanning of a particular 
picture element, applied to the modula- 
tor of the picture transmitter. 

E Y r is the gamma-corrected voltage of 



the monochrome (black-and-white) por- 
tion of the color picture signal, corre- 
sponding to the given picture element. 

ER, EG'I and EB' are tne gamma- 
corrected voltages corresponding to the 
red, green and blue signals intended for 
the color picture tube, during the scan- 
ning of the given picture element. 

E Q ' and EI are the two gamma-cor- 
rected orthogonal components of the 
chrominance signal corresponding, re- 
spectively, to the narrow-band and wide- 
band axes. 

w is 2r times the frequency of the 
chrominance subcarrier. The phase ref- 
erence of this frequency is the color syn- 
chronizing signal (see Specif. 21 below) 
which corresponds to amplitude modula- 
tion of a continuous sine wave of the 
form sin (o>* -j- 180) where t is the time. 

The portion of each expression be- 
tween brackets represents the chromi- 
nance subcarrier signal which carries the 
chrominance information. 

It is recommended that field-test 
receivers incorporate a reserve of 10-db 
gain in the chrominance channel over 
the gain required by the above expres- 
sions. 

11. The primary colors referred to by 
E R ', E G ', and E B r have the following 
chromaticities in the CIE system of speci- 
fication : 

* y 

Red (*) 0.67 0.33 - 

Green (G) 0.21 0.71 

Blue (B) 0.14 0.08 

12. The color signal is so proportioned 
that when the chrominance subcarrier 
vanishes, the chromaticity reproduced 
corresponds to Illuminant C (x = 0.310, 
y = 0.316). 

13. Gamma correction is such that the 
desired pictorial result shall be obtained 
on a display device having a transfer 
gradient (gamma exponent) of 2.75. 
The equipment used shall be capable of 
an overall transfer gradient of unity with 
a display device having a transfer gradi- 
ent of 2.75. The voltages E T ' ', E R r , 
E G ', E B ', E Q ' and / hi the expression 



334 



April 1953 Journal of the SMPTE Vol. 60 




O. 18 H. MAX.- 



Fig. 9. Revised specifications for field test of NTSC 
compatible color television. 



NOTES 

1. The radiated signal envelope shall 
correspond to the modulating signal of 
the above figure, as modified by the 
transmission characteristics of Specif. 6. 

2. The burst frequency shall be the fre- 
quency specified for the chrominance 
subcarrier. The tolerance on the fre- 
quency shall be 0.0003% with a 
maximum rate of change of frequency 
not to exceed -^ cycle/sec/sec. 

3. The horizontal scanning frequency 



shall be 2/455 times the burst frequency. 

4. Burst follows each horizontal pulse, 
but is omitted following the equalizing 
pulses and during the broad vertical 
pulses. 

5. Vertical blanking 0.07 to 0.08 v. 

6. The dimensions specified for the burst 
determine the times of starting and 
stopping the burst, but not its phase. 

7. Dimension P represents the peak-to- 
peak excursion of the luminance signal, 
but does not include the chrominance 
signal. 



of Specif. 1 0, above, refer to the gamma- 
corrected signals. 

14. The color subcarrier frequency is 
3.579545 me 0.0003% with a maxi- 
mum rate of change not to exceed y 1 ^ 
cycle/sec/sec. 

15. The horizontal scanning frequency 
is -f^^ times the color subcarrier fre- 
quency. This corresponds nominally 
to 15,750 cycles/sec (the actual value is 
15,734.264 0.047 cycles/sec). 



16. The bandwidth assigned to the 
monochrome signal E Y ' is in accordance 
with the FCC standard for black-and- 
white transmissions, as noted in Specif. 6 
above. 

17. The bandwidth assigned prior to 
modulation to the color-difference sig- 
nals E Q ' and E/ is given by Table I. 

18. Ey', E R ', E ', E B ', E Q ', and E,' 
are all matched to each other in time to 
within 0.05 /xsec. This is a tentative 



A. V. Loughren: Color Television 



335 



Table I 



Q-Channel bandwidth 

at 400 kc less than 2 db down 
at 500 kc less than 6 db down 
at 600 kc at least 6 db down 

/-Channel bandwidth 

at 1.3 me less than 2 db down 
at 3.6 me at least 20 db down 



tolerance to be established definitely 
later. 

19. The overall transmission band- 
width assigned to the modulated chro- 
minance subcarrier shall extend to at 
least 1.5 me below the chrominance sub- 
carrier frequency and to at least 0.6 



me above the chrominance subcarri* 
frequency, at an attenuation of 2 db. 

20. A sine wave, introduced at those 
terminals of the transmitter which are 
normally fed the color picture signal, 
shall produce a radiated signal having an 
envelope time delay, relative to 0.1 me, 
of zeYo microseconds up to a frequency of 
2.5 me; and then linearly decreasing to 
4.3 me so as to be equal to 0.26 yusec 
at 3.579545 me. The tolerance on all 
these delays shall be 0.05 /zsec relative 
to the delay at 0.1 me. 

21. The color synchronizing signal is 
that specified in Fig. 9. 

22. The field strength measured at any 
frequency beyond the limits of the as- 
signed channel shall be at least 60 db 
below the peak carrier level. 



336 



AprU 1953 Journal of the SMPTE Vol. 60 



Eidophor System of 
Theater Television 

By EARL I. SPONABLE 



The Eidophor, or Fischer, theater television system is described in an intro- 
ductory way, then as installed at the Twentieth Century-Fox home office 
theater a year ago for exhibition shows for exhibitors and the press. 



HE INITIAL CONCEPT of the Eidophor 
System started some thirteen years ago 
when Professor Dr. Fritz Fischer, work- 
ing at the Swiss Federal Polytechnical 
Institute in Zurich, applied for a patent 
on a new and radically different idea 
for obtaining projected television pic- 
tures. His consideration of the problem 
led him to the conclusion similarly 
reached by others that in order to 
have a theater-sized picture of adequate 
brightness it would be necessary to 
employ a high-intensity arc as the light 
source. His thinking was thus along the 
lines of a standard motion picture pro- 
jection unit, and the first effort to con- 
struct such a device, though the result 
was large and cumbersome, followed this 



Presented as an engineering exhibit at the 
theater television hearing before the Federal 
Communications Commission in October 
1952, by Earl I. Sponable, Twentieth 
Century-Fox Film Corp., 460 W. 54 St., 
New York 1 9, N.Y. The engineering exhib- 
its were presented under the joint aus- 
pices of the Motion Picture Association of 
America, Inc., and the National Exhibitors 
Theatre Television Committee. 



general plan. He then reviewed possi- 
ble means for controlling this light beam. 
The modulating units which had pre- 
viously been employed in similar ar- 
rangements by others had not been 
particularly successful, and Prof. Fischer 
conceived the idea of making use of 
optical principles first enunciated by 
Foucault as a means for studying the 
surface configurations of concave tele- 
scope mirrors. These principles had 
later been extended by Toepler, who 
observed the difference in refractive 
index in air caused by heat waves in the 
optical system. Since Toepler's work 
in the early 1860's, this type of optical 
system has been frequently referred to 
as the "Toepler Schlieren" or simply 
"schlieren" system. (Schlieren is a 
German word meaning "streaks" or 
"striae.") An understanding of these 
optical principles is necessary in under- 
standing the Eidophor system, since this 
is the heart of the matter, and it may 
best be explained in terms of the earliest 
of Prof. Fischer's models. While this 
early unit was unsuccessful, it did prove 



April 1953 Journal of the SMPTE Vol.60 



337 



3CRLLN 




Fig. 1. Optical arrangement of schlieren system. 




Fig. 2. Eidophor liquid as diffracting agent at image plane. 



the workability of the basic ideas and 
led to the development of later models 
which performed satisfactorily. 

Figure 1 is a schematic of this optical 
arrangement. The positive crater A of 
the arc lamp together with condenser 
lens B produces uniform illumination of 
the plane C; the light-modulating or 
controlling medium is placed in this 
plane between two bar-and-slit systems 
F and G. A field lens is so arranged 
that it images system F upon the bars 
of system G, at the same time insuring 
correct illumination of each and every 
point on the controlling medium. By 
way of example, the point H is illumi- 
nated in such a manner that the incident 
light beams passing through the slits 
of system F impinge upon the opaque 
bars of system G. The image point H 



is located in the image plane C of the 
objective lens D. This projection lens 
would thus image the point H as point 
H' on a projection screen E. This is 
however impossible since the illuminating 
light beams are blocked off by the bars 
of system G. If, however, the control 
medium located in the image plane G is 
deformed in a suitable manner, diffrac- 
tion of the incident light can be effected, 
and the diffracted parts of the beams may 
be made to pass through the slits of 
system G and reach the projection screen 
as image-forming light. 

Figure 2 shows an oil layer of minute 
thickness at the image plane C (shown 
also in Fig. 1). This layer of liquid, 
called the Eidophor liquid, takes the 
place of the usual motion picture film 
in the film gate, as far as the optical 



338 



April 1953 Journal of the SMPTE Vol. 60 



considerations are concerned. This oil 
layer, by way of illustration, might be 
supported by a glass plate. 

It has been pointed out that the inci- 
dent illumination of every image point 
is blanked off by the strips of the bar 
system shown as G (in Figs. 1 and 2). 
Since every point of the oil layer is 
illuminated and since each light beam 
traversing the layer is prevented, by 
the bars of G, from reaching the lens 
when the oil layer is of uniform thick- 
ness and is homogeneous, then under 
this condition no light will reach the 
screen. There remains then the prob- 
lem of creating an image record in the 
form of some optical inhomogeneity 
which will cause the oil layer, point by 
point, to diffract the light beam through 
the slits of G. This is done by means of 
an electron beam from an electron gun 
which scans the oil layer and forms 
thereon a picture raster. The electron 
gun deposits electric charges point by 
point corresponding to the scanned pic- 
ture and these charges cause minute 
corrugations in the surface of the oil 
layer. Where the surface is corrugated 
as at HI on surface G in Fig. 2, the light 
rays issuing from any such image point 
are diffracted, and part of the light which 
normally illuminates the bars of stop G 
then passes through the slits and produces 
illumination at the conjugate point H'i 
on the projection screen. This illumina- 
tion will become more intense the more 
the Eidophor liquid surface is distorted, 
and we have therefore a means of repro- 
ducing point by point and line by line a 
television picture raster on a full-sized 
screen. 

Figure 3 shows the relation between the 
distribution A of brightness along the 
line of the image to be reproduced and 
the wave-shaped deformation B of the 
surface of the Eidophor layer. The 
deviation of the envelope of the deforma- 
tion curve from the original smooth sur- 
face is proportional to the desired bright- 
ness value. 

The principle of the Eidophor method 




Fig. 3. Relation between distribution 
of image brightness (A) and deformation 
of Eidophor surface (B). 



resides in modulating the cathode beam 
scanning the Eidophor surface by the 
video signal in such a manner that the 
deformations are proportional to the 
instantaneous value of the video signal. 
These wave-shaped deformations are 
brought about by the periodically vary- 
ing distribution of electrical charges as 
stated above. These charges deposited 
on the oil surface by the cathode beam 
cause deformation by means of electro- 
static forces. The wavelength of these 
deformations is always constant and lies 
in the neighborhood of 0.1 mm. The 
height of the waves is proportional to 
the video signal. As the illumination 
of the image points on the projection 
screen is always proportional to the 
height of the waves at the corresponding 
point on the Eidophor, the distribution 
of light over the projection screen corre- 
sponds to the video signal and thus to 
the object to be reproduced. 

The deformation commences at the 
moment in which the cathode beam 
scans a particular point of the image. 
By a suitable choice of the conductivity 
and viscosity of the Eidophor oil, it is 
possible to conserve the deformation for 
a considerable part of the image-scanning 
period so that it disappears but shortly 
before the occurrence of the succeeding 
scanning. In other words, the illumina- 
tion of the projection screen is maintained 
for this part of the scanning period ; this 
represents a substantial light storage 



Earl I. Sponable: Eidophor System 



339 





1. ARC LIGHT SOURCE. 

2. APERTURE PLATE. 

3. COLOR WHEEL. 

4. CONDENSER LENS. 

5. MIRROR BAR SYSTEM. 

6. ELECTRON GUN AND 

DEFLECTION SYSTEM. 

7. SPHERICAL MIRROR WITH 

THIN LAYER OF EIDOPHOR LIQUID. 

8. ELECTRON BOMBARDED LIQUID AREA 

THAT MODULATES LIGHT BEAM. 

9. KNIFE EDGE DETERMINING 

THICKNESS OF LIQUID LAYER. 

10. PROJECTION LENS. 

11. DIRECTING MIRROR. 

12. THEATRE SCREEN. 



Fig. 4. Essential components of the projector. 



effect. The light storage effect is of 
primary importance when considering 
the efficiency of the Eidophor method, 
which is appreciably in excess of the in- 
crease in efficiency which may be 
achieved with ordinary cathode-ray tubes 
by using the afterglow of the phosphor 
layer. 

After the completion of the first proto- 
type model and the study of the data 
gained from it, it was decided to build a 
second and improved unit. Before this 
was finished Dr. Fischer died, and the 
work was carried on by his associates 
under the direction of Professor Baumann 
and Dr. Thiemann. With the second 
unit large-screen images were produced 
and the results were sufficiently promis- 
ing to justify building a third and much 
smaller prototype with a new layout of 
the schlieren optics utilizing only one 
bar system and a spherical mirror. This 
third prototype was first operated in 
December 1950; it showed black-and- 
white, large-screen television pictures 
and fulfilled all expectations as to its 



performance. (A description of the 
early equipment has been published in 
this Journal. 1 Dr. Thiemann published 
an extensive article in 1949, 2 and an 
article by Prof. Baumann 3 is reprinted 
in succeeding pages of this Journal.} 

Since the time had come when the 
emphasis must be on practical design, 
an arrangement was made between the 
Institute and Dr. Edgar Gretener, A.G. 
of Zurich whereby the latter organiza- 
tion would undertake the practical adap- 
tation and commercialization of the 
equipment. Discussions culminating in 
a contractual relationship were nego- 
tiated with Twentieth Century-Fox; 
demonstrations were given, and tech- 
nical conferences held, and it was deter- 
mined that, to compete with the present 
state of the motion picture art, large- 
screen television pictures should be in 
color. 

An arrangement was made with the 
Columbia Broadcasting System to make 
use of its field-sequential color knowledge 
and techniques. This method offers to 



340 



April 1953 Journal of the SMPTE Vol. 60 




Fig. 5. Eidophor large-screen color television projection equipment: (1) Eidophor 
projector; (2) projection light beam hood; (3) color wheel; (4) auxiliary services (vacuum 
pump, thermostat and system for Eidophor cooling); (5) projection lamp (Ventarc-type); 
and (6) television receiver circuits. 



the observer red, blue and green com- 
ponent images with identical contours 
in rapid alternating succession. The 
rapid succession of these three component 
images leads to additive color mixture 
within the eye of the observer, providing 
of course that the alternation of the com- 
ponent images is done with sufficient 
rapidity. 

In order to convert the black-and- 
white Eidophor projector to the field- 
sequential color system, it is only neces- 
sary to insert the usual filter wheel in the 
arc beam at a point conjugate to the 
image plane, and then to cause the filter 
wheel to rotate in synchronism with a 
similar wheel in the camera. If the 
appropriate color signals are then fed 
to the electron gun, appropriate colored 
images will be projected to the screen. 
Since the field-sequential color system 
has already been approved by the Com- 
mission and has been the subject of exten- 
sive prior exhibits, 4 it seems unnecessary 
to describe it more fully here. A sche- 
matic drawing of the arrangement of the 



essential components of the projector is 
shown in Fig. 4. 

The third prototype unit was modified 
for color and operated in Zurich in 
September 1951, and was shown Novem- 
ber 12 to an audience of some 55 per- 
sons consisting of representatives of 
motion picture exhibitors, industry execu- 
tives, and the interested press. This 
equipment was subsequently shipped to 
New York and installed in the home office 
theatre of Twentieth Century-Fox, where 
it was ready for operation in April 1952. 
Although still an engineering model, its 
performance was most gratifying and 
many demonstrations were given with it. 
For this particular unit the technical 
standards were as follows: 

Field repetition rate 150/sec 

Frame repetition rate 75 /sec 

Interlace 2/1 

Color picture repeti- 
tion rate 25/sec 

Number scanning lines 

(total) 525 



Earl I. Sponable: Eidophor System 



341 





HOME OFFICE THEATER 
t- + t- WEST 56THST. 




MOVILTONL 

50 IVE5T J> Ttl 3T 



Fig. 6. Typical complete Eidophor system. 



Horizontal scanning 
frequency 

Horizontal blanking 

Vertical blanking 

Horizontal definition 
(system) 

Video bandpass (sys- 
tem) 

Field color sequence 

Highlight screen 
brightness 

Screen size 

Throw 



39,375/sec 
20% (approx.) 
10% (approx.) 

400 lines (approx.) 

12 me (approx.) 
red, blue, green 

6 ft-L (approx.) 
11 X 15ft 
68ft 



Figure 5 shows the experimental 
Eidophor equipment as set up in New 
York, and shows particularly how the 
auxiliary equipment (required to main- 
tain the Eidophor liquid at constant 
temperature, and the cathode-ray system 
under constant low pressure) has been 
integrated into one compact unit. Ap- 



propriate automatic controls and protec- 
tive devices have also been incorporated 
into the system, thus making the equip- 
ment comparable to an ordinary motion 
picture projector in its operational facil- 
ity. 

In order to give a more comprehensive 
picture of a typical complete system, a 
schematic diagram is shown in Fig. 6 
of the arrangement of the various func- 
tional components used for demonstra- 
tion purposes. This diagram is illustra- 
tive of a situation in which a studio 
sends the video signal to the theater, or 
theaters, over suitable transmission cables 
or radio relay links. 

It may be appropriate to point out 
here that the Eidophor system is by no 
means limited to field-sequential color. 
It has been suggested (and development 



342 



April 1953 Journal of the SMPTE Vol. 60 



work is being done in this direction) that 
several picture rasters be laid down in 
appropriate geometric relationship on 
the one mirror surface, each raster corre- 
sponding to one of the primary colors 
chosen for the particular process. Thus 
either a two-, three-, or four-color addi- 
tive simultaneous projection process is 
theoretically possible, with consequent 
economy of light, but with obvious com- 
plication in equipment design. 

The development of the Eidophor 
system has now reached a point where 
the quality of reproduction can be fairly 
judged. As a result of observations of 
this system in operation and of the favora- 
ble comments received from others, plans 



are now being made for the development 
and production of a commercial model. 

References 

1. E. Labin, "The Eidophor method for 
theater television," Jour. SMPTE, 54: 
393-406, Apr. 1950. 

2. H. Thiemann, "Large-screen television 
projection with the Eidophor system," 
Bulletin of the SEV, No. 17, 1949, Swiss 
Electro-Technical Society (in German). 

3. E. Baumann, "The Fischer large-screen 
projection system," J. Brit. IRE, 72 
(new series): 69-78, Feb. 1952. 

4. P. C. Goldmark, J. W. Christensen and 
J. J. Reeves, "Color television 
U.S.A. Standard," Jour. SMPTE, 57: 
336-381, Oct. 1951. 



Earl I. Sponable: Eidophor System 



343 



The Fischer Large-Screen 
Projection System 



By E. BAUMANN 



The paper describes the Fischer system of large-screen television projection 
which makes use of an "Eidophor" liquid. Consideration is given to the 
brightness requirements, and the principles of the schlieren optical system 
are discussed together with the manner in which the Eidophor functions. 
The technical problems involved are dealt with and the results obtained with 
the third prototype examined. 



1. Introduction 

There are many possibilities for the 
solution of the difficult problem of 
large-screen projection of television pic- 
tures, and it would almost be foolish to 
consider one or the other method as 
absolutely best. The system which is 
to be discussed here has been influenced 
from its beginnings by a clear and defi- 
nite aim: the television picture should, 
in every respect, be of the same quality 
as the modern motion picture, i.e., of 
equal brightness, size, definition and 
gradation. I realize of course that this 
is a very ambitious aim and I would by- 
no means maintain that it has already 
been reached at present. On the other 



Presented at the Fifth Session of the 1951 
Radio Convention on August 25 in the 
Cavendish Laboratory, Cambridge, Eng- 
land, by Prof. E. Baumann, Institut fur tech- 
nische Physik an der E.T.H., Gloriastrasse 
35, Zurich 7, Switzerland. Reprinted from 
the Journal of the British Institution of Radio 
Engineers, 12 (new series}: 69-78, Feb. 1952. 



hand, I should like to make it clear that 
all our work has always been done with 
this aim in view. And as the develop- 
ment proceeds, the fundamental diffi- 
culties that might hinder the final suc- 
cess grow smaller and smaller. 

You will allow me a short summing up 
of how the system has been developed so 
far. Professor Dr. F. Fischer applied 
for a patent for his basic ideas in Novem- 
ber, 1939, and started their practical 
development in the AFIF (Department 
of Applied Physics of the Swiss Federal 
Institute of Technology, Section of 
Industrial Research). With his almost 
unlimited courage he tackled the under- 
lying problems that had never been 
treated before. He constructed the 
first prototype of the projector which 
was first operated in the year 1 944. The 
results of this construction, however, 
were by no means encouraging and did 
not by far fulfil the hopes that were put 
in it. The difficulties, it is true, were 
not of such nature as to render any fur- 
ther pursuit of the idea absolutely hope- 



344 



April 195* Journal of the SMPTE Vol.60 



Fig. 1. Principle of pro- 
jection. The projection 
lens of diameter D is 
at a distance from S of 
about / = focal length 
(for lateral magnifications 
of m = \/~A'IA > 1). 




less, but it seemed clear that the success 
could only be reached at a much greater 
expense of research and development 
than was originally thought. But Prof. 
Fischer did not allow himself to be 
troubled and courageously started to 
design a second prototype. Already 
the outward appearance of the second 
projector made it clear that his bold 
original plans had been revised. Still 
the equipment was very large and com- 
plicated and there could hardly be any 
serious hope of its practical adoption in 
that shape. But the first target that 
had to be reached was the proof that 
the system could work at all. When 
Prof. Fischer's untimely death in the 
year 1947 brought his tireless activity 
to an end, the second projector was not 
yet in operation and the main problems 
were by no means solved. We do not 
lessen Prof. Fischer's lasting merits as 
a pioneer, if we say that only the develop- 
ment made after his death gave the 
system a basis for its practical adoption. 
The success is in the first place due to 
the optimism and untiring energy of his 
former collaborators, Dr. H Thiemann, 
F. Mast, K. Hetzel, Dr. F. Held and Dr. 
R. Petermann. 

On the occasion of the International 
Television Convention in the summer of 
1 948, the operation of the second proto- 
type could be demonstrated. The qual- 
ity of the pictures which was reached 



then gave proof of considerable progress, 
and helped to strengthen very widely 
the hope of final success. Those respon- 
sible were convinced that the new ex- 
periences and discoveries would furnish 
the essential data for a third prototype 
which, considerably simpler and smaller 
than the second, should come much 
nearer to a solution that could be prac- 
tically adopted. This third prototype 
which was first operated in December, 
1 950, fulfilled our hopes in every essential 
point. 

Since in this stage the stress lay on the 
problems of practical adaptation, it was 
time to secure the help of a competent 
industrial producer and to leave to him 
the responsibility for this difficult task. 
This was done in the spring of this year 
and the rights for the utilization of our 
system now lie with the firm Dr. E. 
Gretener A.G., in Zurich. As a univer- 
sity department we have retained the 
possibility to further the development in 
close collaboration with the firm. 

2. Theoretical Requirements 
of the System 

It will be known that the Eidophor 
system makes use of a light source for the 
production of the screen picture. The 
brightness, therefore, as in cinema pro- 
jection, depends largely on the power of 
this light source. This fact and the 
relatively great efficiency of the system 



E. Baumann: Fischer System A Reprint 



345 



Table I. Screen Areas and Light Fluxes in Some Cinemas on the Continent 



Cinema 


Seats 


Screen 
size 
(picture 
width) Screen area, 
in ft sq m 


Projec- 
tion 
dis- 
tance, 
m 


Projec- 
tion 
angle, 
deg 


Light 
flux 
(ASA), 
1m 


Kapitol in Bern 


720 


19 


4.2 X 5.6 


= 23.5 


34 


17 


3161 


Urban in Zurich 


1150 


22 


5 X 6.7 


= 33.5 


31.5 


12 


4506 


Rex in Paris 


3200 


33 


7.5 X 10 


= 75 


45 


21 


10087 


Marignan in Paris 


1800 


40 


9 X 12 


= 108 


33.5 


23 


14526 


Le Regent in Neuilly- 
















sur-Seine 


1400 


19 


4.2 X 5.6 


= 23.5 


26 


15 


3161 


Cineac des Ternes in 
















Paris 


600 


19 


4.2 X 5.6 


= 23.5 


22 


18 


3161 


Cineac in Strasbourg 


500 


16 


3.5 X 4.7 


= 16.5 


21 


23 


2219 


Royal in Rabat 


1500 


20 


4.5 X 6 


= 27 


39 


20 


3631 


Palazzo del Cinema in 
















Venice 


1400 


33 


7.5 X 10 


= 75 


34 


12 


10087 


Palladium in Copen- 
















hagen 


1400 


27 


6X8 


= 48 


35 


5 


6466 


Asta at The Hague 


1200 


22 


5 X 6.7 


= 33.5 


27 


3 


4506 


Capitol at The Hague 


1000 


20 


4.5 X 6 


= 27 


23.5 


17 


3631 



enable us to produce very powerful light 
fluxes which are sufficient even for the 
screens of the largest theatres. A fur- 
ther advantage lies in the possibility of 
adapting the optical layout of the system 
to the architectural necessities of the 
various theatres and to install the pro- 
jector in the existing booth, as it is com- 
monly done with ordinary cinema equip- 
ments. 

The brightness B s of the screen re- 
quired in good cinema projection amounts 
to about 10 ft-L. For a given screen 
area A' the required light flux can be 
computed as follows 



- 078 A>B > 



(1) 



The factor 0.8 is the coefficient of remis- 
sion of the screen, as for an ordinary 
medium quality cinema screen with dif- 
fuse reflexion. 

Table I states the screen areas used in 
some well-known cinemas on the Con- 
tinent. The light flux necessary for these 
screens according to Eq. 1 is given as 
well. The table shows that 0, = 10,000 
1m are necessary for the largest theatres. 



This means an enormous light-flux in 
television, and we have now to consider 
how it can be produced. 

We shall assume, without going into 
details now, that the projection of our 
picture follows exactly the same prin- 
ciples as cinematographic projection. 
Only we have the Eidophor surface 
instead of the film and we shall therefore 
speak of the Eidophor picture. In 
Fig. 1 the Eidophor picture of the area A 
lies in the plane S, the projection screen 
of the area A' in the plane S r . The 
total light flux that falls on the screen is 
equal to the light flux passing the lens. 
From every point of the Eidophor pic- 
ture a light-pencil of half the opening 
angle, 2 0, hits the lens; thus the light flux 
can be computed to be 

(2) 



B being the brightness of the Eidophor 
picture, and sin 6 being ~ D/2f, 



(3) 



(D/f)~ l is called the /-number of the 
lens. 



346 



April 1953 Journal of the SMPTE Vol.60 



If we put the expression from (1) for 
, and using m = \/A'/A we get: 



From (4) we can compute the necessary 
brightness B of the Eidophor picture, if 
we know the brightness of the screen. 
If the screen area A' = 75 sq m = 7.5 X 
10 sq m and the Eidophor picture area 
A = 7.5 X 10 sq cm = 75 sq cm, we 
get m = 100. In cinema projection the 
area A has a fixed value and is much 
smaller namely 3.5 sq cm, which makes 
m = 463. Table II presents the values 
of B for some values of D/f for both 
cinema and television. 



Table II 





m = 100, 


m = 462, 


D/f 


B 


B 


1:1 


175 


3,740 


1:2 


700 


15,000 


1:5.5 


5,250 


114,000 



' B in candles /sq cm. 

It follows that the area A plays a 
decisive part, and if we compare with 
cinema projection we see that in television 
we have the advantage of having this 
area 20 to 30 times larger. In conse- 
quence, we can use apertures (of about 
1 :5.5) which do not present great optical 
difficulties even in more complicated 
systems, and which allow quite reasonable 
lens diameters. However, the projec- 
tion cannot be effected without losses, 
as we have assumed so far. The losses 
that we have to take into account are 
very considerable and may reach the 
factor of 10 (and more) against about 
three for the cinema. The brightness 
of the Eidophor picture actually required 
for the production of 10,000 1m is, 
therefore, in our example for an //num- 
ber of 5.5 about 50,000 to 70,000 c/sq cm. 

But modern arc lamps, such as the 
Ventarc lamp developed by the firm 



Dr. E. Gretener, A.G., can produce 
average brilliancies in the crater up to 
and above 120,000 c/sq cm, so that the 
conditions in this respect are fairly 
favourable. 

I should like to point out that the 
figures mentioned here refer to one single 
example. But according to the given 
practical requirements, all sorts of varia- 
tions can be made. The only purpose 
of the figures given above is to show that 
the production of light fluxes as required 
in the largest theatres can be effected 
by existing means. 

3. Description of the Optical System 

The basis of our projection system is a 
kind of light valve and we shall presently 
see the elegant principles that govern 
the light control. The most important 
element is the so-called schlieren-optical 
system. Figure 2 shows the diagram of 
such a system. The image of a system 
of bars and slits, lying in the plane S and 
illuminated from the left, is projected 
by the schlieren-optical system O s on to 
an analogous system of bars and slits 
lying in S'. As shown in our diagram, 
the images of the slits 7 and 2 fall on 
the bars T and 2', and the images of the 
bars, a, b, c, on the slits a', b r , c' . Under 
these circumstances no light is allowed 
to pass into the space to the right of S' '. 
We now expand the system by joining a 
new lens P immediately behind .S" 
(Fig. 3) and choose its focal length so as 
to project the image of a plane P near 
the schlieren lens O s , on to the screen P'. 

A small glass prism N is now placed 
on a point of the plane P. Conse- 
quently the light-pencil, passing N is de- 
flected from its original direction accord- 
ing to the deflection angle of N. The rays 
of this pencil can therefore partly pass the 
bars T and 2' and are focused by the 
lens Op on the screen in N' t where they 
appear as a luminous point. The bright- 
ness of the point N f depends on how 
much of the light-pencil is allowed to 
pass bars 7', 2', and this varies according 
to the deflection angle of the prism. We 



E. Baumann: Fischer System A Reprint 



347 




Fig. 2. Principle of a 
schlieren optical system. 




N' 



Fig. 3. Development 
of a schlieren optical 

system. 



BLACK 
SURFACE 



WHITE 
SURFACE 




BLACK-WHITE 
SURFACE 



Fig. 4. Deformation of the Eidophor 
liquid. 

now imagine the whole plane P to be 
covered with as many adjustable prisms 
as there are picture points on the pic- 
ture to be projected. We can now 
control the brightness of any image point 
on the screen P' by suitably adjusting 
the deflection angles of the prisms. Our 
pictures are thus produced on a raster 



basis, i.e., they are composed of individual 
discrete picture points. 

It was Prof. Fischer's ingenious idea to 
produce the prisms by depositing a ras- 
tered electric charge on a thin oil film. 
Under the influence of the electrostatic 
forces the surface of the oil is deformed 
and in this way the required raster ele- 
ments are formed. The exact geomet- 
rical shape of these elements is important 
only in so far as it influences the light 
efficiency to a certain extent. The 
charge is applied point by point by means 
of a cathode beam, and we can thus 
employ the methods used in the tech- 
nique of cathode-ray tubes. In order 
to produce a television picture, deforma- 
tions in the form of sine-shaped wave- 
traces are engraved along the lines of the 
picture. The amplitude of the wave 
controls the brightness of the correspond- 



348 



April 1953 Journal of the SMPTE Vol. 60 




Fig. 5. Diagrammatic arrangement of the 

"Eidophor" projector. 
(Based on the second prototype.) 



(a) Electron gun 

(b) Modulation grid 

(c) Focusing coil 

(d) Scanning coils 

(e) Arc lamp 



(f) Lower grating 

(g) Eidophor 

(h) Upper grating 

(i) Objective lens 

(j) Screen 



ing part of the screen picture. In Fig. 
4 the system is shown schematically. 

An ideal reproduction equipment 
should be able to store the picture infor- 
mation for the duration of one picture 
period. This means that the deforma- 
tions on the oil should remain for the 
duration of one picture period, but decay 
as quickly as possible after the period is 
over. This decay, however, is only 
possible if the electric charge that pro- 
duces the mechanical forces decays as 
well. For this purpose, the oil, the 



Fig. 6. (a) Build-up of the deformation 
for a nonconductive liquid. 



(b) Decay of the charge in a 
conductive liquid. 




(c) Development of the deformation in 
a conductive liquid. 



Eidophor as Prof. Fischer has called it, 
has been made conductive. The de- 
posited charges, therefore, decay accord- 
ing to a time law corresponding to an 
exponential function. The electric con- 
ductivity governs the time constant. 
The mechanical processes of the forma- 
tion and the decay of the deformation 
are subject to a certain inertia. The 
decisive forces for this are the surface 
tension which tends to level out the sur- 
face of the liquid, and the viscosity 



E. Baumann: Fischer System A Reprint 



349 




Fig. 7. Development of the deformation of the medium with time. 



which takes the function of damping. 
If we watch one given image point we 
observe the following process: At the 
beginning of the image period, the cath- 
ode ray passing the image point deposits 
an electrical charge within a very short 
time < 10~ 7 sec. The electrostatic 
forces become effective at once and the 
deformation of the surface begins accord- 
ing to an exponential law (cf. Fig. 6). 
The time constant is proportional to the 
quotient of the surface tension and the 
viscosity. For a nonconductive Eido- 
phor the deformation would tend to 
reach a definite final value (Fig. 6a). 
As, however, the charge diminishes 
according to an exponential function, 
the forces causing the deformation are 
diminished as well (Fig. 6b), so that 
both causes result in a development of 
the deformation as illustrated in Fig. 6c. 
For a whole series of successive image 
periods, a development as illustrated in 
Fig. 7 results. For all practical purposes, 
a remainder of 10% of the deformation, 
at the end of the image period, is tolera- 
able. This time law of the deformation 
allows an effective light-storage of about 
70%. _ 

As is easily seen, the Eidophor image 
constantly carries a certain average nega- 
tive charge which exercises a constant 
average mechanical pressure on the oil 
surface. If the oil film were left to itself 
it would be pushed out of the image 
field in the course of time. To prevent 
this, the Eidophor carrier is slowly 



being turned so that the oil film in the 
image field is constantly renewed. 
Since the rotation is very slow, its influ- 
ence on the image is practically nil. 

The maximum amplitude of the de- 
formation is only a few thousandths of 
a millimeter. Consequently, the oil 
surface must be of the highest quality 
as even the smallest deficiencies result 
in an undesirable brightening-up of 
the screen. Fortunately this trouble 
can be removed to a certain extent in a 
comparatively easy way. If the relation- 
ship between the dimension of the raster 
elements and the other dimensions of 
the schlieren-optical system is appro- 
priately chosen, the deflection of the 
light is accompanied by diffraction, 
which allows for a schlieren system of 
lower imaging quality to be used success- 
fully. 

We can now make the second bar 
system 7', 2' (Fig. 8) somewhat wider 
than the optical image of the slits 7, 2. 
The result is an elimination of the bright- 
ening-up effect of certain deformations, 
deformations exceeding a certain area 
now remaining without effect. 

This new layout of the schlieren-op- 
tical system has made it possible to design 
a new simplified projector which is also 
reduced in size. Its diagram is shown 
in Fig. 9. The schlieren lens was re- 
placed by a spherical mirror. Instead 
of two separate bar systems we now use a 
single one which comes twice in action. 
It consists of a number of mirror strips 



350 



April 1953 Journal of the SMPTE Vol. 60 



Fig. 8. Schlieren systei 
using diffraction. 




Fig. 9. Diagram of the Eidophor projector (third prototype). 



which are fixed at twice the focal dis- 
tance, at an angle of 45 with respect 
to the optical axis of the mirror. Thus 
the bar system carries its own image that 
is reflected by the mirror. The Eidophor 
picture itself is placed on the spherical 
mirror and again consists of a thin oil 
film. The mirror is now caused to 
rotate slowly and a straight radial ruler 
allows the passage of a quantity of oil 
necessary for the production of the picture 
carrier. 

Optically the system works as follows: 
The mirror strips are illuminated by an 



arc lamp, the light-pencil of which is at 
a right angle to the axis of the spherical 
mirror. By an illuminating lens an 
image of the arc crater is formed on the 
Eidophor. The relative position of the 
bar system and the spherical mirror is 
so as to cause all the light to fall back on 
the arc, owing to the focusing power of 
the spherical mirror, as long as the Eido- 
phor surface is absolutely smooth. But 
as soon as the Eidophor surface is de- 
formed, the light is accordingly allowed 
to pass through the slits of the bar system 
and to enter the projection lens. It 



E. Baumann: Fischer System A Reprint 



351 




Fig. 10 (above). 
View of third proto- 
type projector. The 
"electron gun" struc- 
ture can be clearly 
seen, also the Eido- 
phor container. 



(Right) View of the 
projector from the 
rear. The arc lamp 
projector is promi- 
nent, and the high- 
vacuum equipment 
can be seen below it. 




352 



April 1953 Journal of the SMPTE Vol. 60 



then reaches the screen by a deflecting 
mirror. Of course, full use is made of 
the diffraction optic and the edges of the 
mirror strips are, therefore, blackened 
and nonreflecting. The deposition of 
the charge on the oil film has been taken 
over from the old system without any 
change. 

The principal advantages of the new 
system as compared with the old one are 
the following. Owing to the use of a 
reflecting optic, the light path through 
the schlieren system is now only half 
its original length. The diffraction optic 
also allows us to make use of a large 
image angle in the schlieren system and a 
consequent small focal length of the 
mirror. This leads to a much more con- 
venient optical layout and to a considera- 
ble reduction of the outward size of the 
projector. Even now it does not take up 
more space than an ordinary cinema pro- 
jector. This reduction in size is clearly 
illustrated by the fact that the distance 
of the two bar systems of the schlieren- 
optical system was about 3 m in the old 
projector whereas the distance between 
the reflecting bars and the mirror is 
only 75 cm in the new system. Also 
the cooling of the Eidophor liquid has 
been made easier, as the Eidophor pic- 
ture itself can now be cooled. This, in 
turn, allows a reduction of the rotating 
speed of the mirror so that the produc- 
tion of interlaced pictures is possible 
without difficulties. 

4. The Definition of the System 

Having seen the underlying principles 
of the operation of this system we shall 
now discuss some of the most important 
details. A problem of the first impor- 
tance is, of course, the question of defini- 
tion. Contrary to certain other systems, 
the definition of the picture is totally 
independent of the light-capacity, which, 
of course, is a great advantage. We 
know that the pictures are rastered. The 
attainable limit of the resolving power is 
dependent on the number of the raster 
elements and this might be the cause of 



a certain limitation. But this is, in 
practice, by no means the case as it is 
possible to produce more than 1000 
raster elements per line, a number which 
meets all requirements for the time being. 
The same can be said about the number 
of lines. We did not limit ourselves 
to any fixed number, and our system 
has been working at 392, 625, and 729 
lines. The situation may be most 
clearly illustrated by saying that the 
system allows frequency bandwidths of 
10 me and more to have their entire pic- 
ture information on the screen. 

5. Production of Picture Raster 

Contrary to the functioning of the 
ordinary cathode-ray tubes we do not 
use amplitude modulation but a kind of 
velocity modulation of the cathode ray. 
To produce the picture raster we have 
to deposit a periodic distribution of the 
charge along every line of the Eidophor 
picture, the magnitude of this charge 
being proportional to the brightness of 
every picture point. 

The cathode ray which is of constant 
intensity shows a rectangular cross sec- 
tion on the oil surface. The height of 
this rectangle is the same as the width of 
one picture line, its width being about 
10 to 20% of the height. As long as the 
cathode ray travels along the line with 
constant velocity, it will deposit per unit 
of length a constant charge on the sur- 
face, the density of this charge being 
proportional to the writing speed for a 
constant intensity of the beam. We 
can thus influence the density of the 
charge by varying the writing speed. 
If the speed is great, a small charge is 
deposited; if it is small the charge be- 
comes proportionately greater. In order 
to produce the modulation, an alternat- 
ing voltage of constant frequency is 
superimposed on the line-sweep voltage. 
The frequency of this additional volt- 
age controls the dimension of the raster 
elements, the amplitude controlling 
the density of the charge deposited (cf. 



E. Baumann: Fischer System A Reprint 



353 



CHARGE 
DENSITY 





Fig. 11. Velocity modulation of the 
cathode ray. 

Fig. 11). For the practical applica- 
tion of this modulating system it is useful 
to introduce the additional modulation 
velocity through separate plates. For 
the full control of the picture modulation, 
potentials of about 1 v are sufficient. 

It can be seen from the way the raster 
is produced that it is very important 
that the size and shape of the cathode- 
ray spot remains unchanged over the 
whole picture area. Great care must 
therefore be given to an accurate focusing 
of the beam. Any widening of the spot 
causes the picture to grow darker, but the 
definition of the picture is hardly influ- 
enced by this. In order to obtain a con- 
stant rectangular shape of the spot, a 
mechanical tungsten diaphragm is placed 
in the cathode-ray optic in the crossover 
of the electrons in front of the cathode; 
the rectangular opening of the diaphragm 
is 0.1 X 0.015 mm (= 0.004 in. X 
0.0006 in.). A magnetic lens projects 
the image of this opening electronically 
on the Eidophor surface at the scale 1:1. 
The considerable inclination of the 
cathode ray with respect to the picture 
plane demands, of course, adequate cor- 
recting potentials, which must be varied 
with great precision so as to assure good 
focusing. Moreover the ordinary diffi- 
culties that arise in any electron-optical 
system have to be overcome, and the 
actual limitation of the imaging quality 
of the crossover diaphragm lies in the 
aperture defects of the magnetic lens. 

A valuable help in the solution of the 
whole problem is of course the constant 



beam current. In order to obtain the 
necessary density in the cathode beam, 
a special cathode was developed. Un- 
fortunately the use of oxide-coated 
cathodes is impossible. The hydro- 
carbon vapours produced by the Eido- 
phor' would soon render such a cathode 
ineffective. For the time being pure 
tungsten is used as material for the cath- 
ode. Figure 1 2 shows a diagram of the 
cathode now used. The carrier of the 
emission is a tungsten bar which tapers 
at its back end to avoid a great loss of 
heat. This bolt is surrounded by a helix 
made of tungsten wire which serves to 
heat up the bar. The transmission of 
heat through the radiation of the heated 
helix does not suffice to bring the bolt 
to the high temperature required. We 
therefore apply to the helix a potential 
of adequate magnitude, negative with 
respect to the bolt. The bolt is thus 
submitted to an intense bombardment 
of electrons, which results in an addi- 
tional heating up. By means of this 
device, for the size of the spot as given 
above, the beam current can reach about 
20;uA, an intensity which is sufficient 
for the full control of the television pic- 
ture for line numbers and picture fre- 
quencies as they are used today. The 
picture and line deflections are effected 
magnetically. The dimensioning of the 
deflection elements was given special 
care, not only in connection with the 
problem of focusing, but also in order to 
reduce the distortion of the picture to a 
minimum. 

6. The Eidophor Liquid 

The technological problems that had 
to be solved in the development of the 
whole system were numerous and also 
very difficult in parts. The Eidophor 
liquid and the cathode beam are placed 
in a vacuum. In order to keep the 
pressure at about 10~ 5 to 10~ 4 mm Hg, 
a continuously operating oil-diffusion 
pump is mounted on the projector. 
Vacuum technique is so easily handled 
today that such a pump no longer pre- 



354 



April 1953 Journal of the SMPTE Vol. 60 



sents any serious difficulties for the prac- 
tical application of the projector. 

A problem of fundamental importance 
was the production of a suitable Eidophor 
liquid because it had to meet require- 
ments in many respects. In the begin- 
ning of the development it was a disputa- 
ble point whether or not the oil would 
bear the intense bombardment of elec- 
trons of a velocity of about 20 kv without 
being destroyed. Indeed, some initial 
difficulties in this respect had to be over- 
come, but we are now sure that the Eido- 
phor liquid can have a quite considerable 
life. If the oil was suitably chosen, no 
perceptible destruction was observed. 
Further qualities of the Eidophor liquid 
should be: very small vapour pressure 
(less than 10~ 5 mm Hg) optimal surface 
tension, dielectric constant, viscosity and 
electric conductivity. Of course the 
Eidophor should be as transparent as 
possible to prevent any influence on the 
colour of the screen picture. It is thus 
easily seen that the production of a 
suitable Eidophor liquid is a science of its 
own. Although many improvements are 
still to be desired, the results reached so 
far are quite satisfactory, especially if we 
consider the complicated nature of the 
problem. 

It was mentioned before that the 
decay of the Eidophor deformation 
depends on the viscosity of the oil. As 
the viscosity, in turn, is largely dependent 
on the temperature, it is necessary to 
keep the temperature of the Eidophor 
constant during the period of its opera- 
tion. This constancy however need not 
be very accurate, and an ordinary refrig- 
erating machine, such as is commonly 
used in household refrigerators, can be 
used for this purpose. 

The whole process of the production of 
these images is a very sensitive method 
to show very small optical inhomogene- 
ities. Unfortunately there are not only 
the voluntarily produced inhomogeneities 
which serve for the production of the pic- 
ture; the oil surface sometimes also 
shows undesired ones that appear as 




Fig. 12. Cross section of the cathode 
of the electron gun. 



defects of the screen picture. Our worst 
enemy in this respect is any kind of dust 
that may enter whilst the system is put 
together and which may then be de- 
posited on the optically sensitive parts. 
One of the most critical points is the 
spherical mirror which is at the same time 
the carrier of the Eidophor. The polish 
of its glass surface must be of an excellent 
quality and the metallizing (done by 
evaporation) can hardly be carried out 
carefully enough. 

7. Conclusions 

The results that have so far been 
reached with the third prototype can 
be described as follows: The laboratory 



E. Baumann: Fischer System A Reprint 



355 



equipment is capable of projecting 4000 
to 4500 1m on the screen. The contrast 
ratio between "black screen" and "white 
screen" is 1:200 to 1:400. The pro- 
ducible contrast in a picture is naturally 
largely influenced by the projection 
room, since the quantity of ambient 
light brightening up the dark parts of 
the picture depends on the nature of this 
room. The subjective impression that is 
given by the picture can quite well be 
compared with that of an ordinary motion 
picture. Gradation and definition will 
of course vary according to the quality of 
the picture source. Experience has 
shown that the observer has a tendency 
to judge the gradation of large pictures 
much more severely than the one of an 
ordinary home set. A very careful elec- 
trical correction is therefore necessary. 
With respect to definition no final verdict 
can yet be given, but interlaced pictures 
of 625 lines have yielded quite encourag- 
ing results. It is certain that any limits 
of the quality of the picture lie for the 
time being in the picture source and not 
in the projector. If there have been any 
troubles that could not be eliminated so 
far they have been due mostly to dust 
particles that could not be removed in 
time. But the continual progress ob- 
served justifies an optimistic outlook. 



Our system will have to stand the 
practical test outside the laboratory 
before long. We know very well that 
the final practical success is still far away 
and we are far from underestimating the 
difficulties. But we also have every 
reason for looking ahead optimistically, 
hoping to have that bit of luck which is 
indispensable for any kind of undertak- 
ing. 

Bibliography 

1. F. Fischer and H. Thiemann, "Theoreti- 
cal considerations on a new method of 
large-screen television projection," 
Schweiz. Arch, angew. Wiss. Tech., 7 (Nos. 

1, 2, 77, 72): 1941; 8 (Nos. 7, 5, 6, 7, 

10} : 1942. (In German) 

2. H. Thiemann, "Theoretical studies of 
the use of quasi-insulating Eidophors 
for large-screen television projection," 
Schweiz. Arch, angew. Wiss. Tech., 73: 
147-154, 178-182, 210-217, May-Aug. 
1947. 

3. H. Thiemann, "Large-screen television 
and the Eidophor process," Onde Elec., 
28: 409-411, Oct. 1948. (In French) 

4. H. Thiemann, "Large-screen television 
and the Eidophor system," Telev. Franc., 
50: 6-10, 1949. 

5. E. Labin, "The Eidophor method for 
theater television," Jour. SMPTE, 54: 
393-406, Apr. 1950. 



356 



April 1953 Journal of the SMPTE Vol. 60 



Review of Work on Dichroic Mirrors and 
Their Light-Dividing Characteristics 

By MARY ELLEN WIDDOP 



During World War Two, the application of thin films to glass surfaces pro- 
gressed from experimental laboratory work to practical application in optical 
devices. At present, virtually all lenses are coated to reduce reflection. Plates 
coated with multilayer films, known as dichroic mirrors, are also being applied 
as efficient light dividers. The recent use of dichroic mirrors in experimental 
color television equipment is one of the best known applications. The purpose 
of this paper is to review briefly the interference phenomenon involved and to 
describe the results obtainable with several dichroic mirror designs. 



X~\BOUT FIFTEEN years ago scientists 
began to experiment with the application 
of very thin multilayer films to glass, in 
order to reproduce the effects of color 
selective reflection and transmission 
strikingly exhibited in nature by certain 
crystals. Since that time, the use of 
this method to produce mirrors which 
efficiently divide light has become well 
known. One of the first applications 
of interference mirrors was in photocell 
monitoring of sound recording for motion 
pictures. During World War Two, 
quantities of dichroic interference mirrors 
were used in range finders and radar 
cameras. They are now used in both 
military and civilian products. In 
the RCA compatible system of color 
television, dichroic mirrors are used to 
divide light into primary colors and to 
combine the primary colors for monitor- 
ing. 

Like the colors of soap bubbles, thin 
films of oil on water, oxide films on 



Presented on October 7, 1952, at the Soci- 
ety's Convention at Washington, D.G., by 
Mary Ellen Widdop, Radio Corporation of 
America, RCA Victor Div., Engineering 
Products Dept., Camden 2, N.J. 



heated steel and certain crystals such as 
chlorate of potash, the light-dividing 
characteristics of dichroic mirrors result 
from interference in reflected and trans- 
mitted light. Interference films are 
also used to produce neutral light divi- 
ders, or to reduce reflection from glass. 
The films are applied to glass by the 
deposition of vaporized material on the 
glass surface, in a vacuum. The thick- 
ness of a deposited film may be controlled 
by observing the changes in intensity of a 
beam of light reflected from the glass, 
through a filter, into a photocell. The 
intensity fluctuates, reaching a maximum 
or minimum whenever the thickness of 
the deposited film is effectively equal to 
a multiple of J of the wavelength of the 
control light. 

To review the interference effects of 
thin films, we will first consider the effects 
produced by single thin films, as shown 
in Fig. 1. These diagrams represent 
plane parallel, nonabsorbing films on 
glass surfaces, and their effect on an 
incident light ray. The film thickness 
in both cases is effectively equal to J of 
the wavelength of the incident light ray. 
The film in diagram (a) has an index of 



April 1953 Journal of the SMPTE Vol 60 



357 



x=x 



A B 1 C' D 1 E" F AX A" B' C D' E' F"_ 




W\A/V I 
Y\ \ \ \ T 
\ \ \ \ \ 



N HIGH 




\\ \ \ \ \ 
V \ \ \ \ 

Vvvvx 



ABODE 



A B C D E 



(A) (B) 

Fig. 1. Interference effects in single 4 X films on glass. 



refraction higher than that of glass, while 
the film in diagram (b) has an index of 
refraction lower than that of glass. 

Any light ray incident on such a film 
becomes the source of two sets of almost 
superimposed parallel rays. These rays 
tend to cancel or reinforce each other by 
destructive or constructive interference. 
The type of interference which occurs 
depends on the relative phase of the 
interfering rays. The phase of the 
reflected and transmitted rays is deter- 
mined by the optical path through the 
film and by phase reversals which occur 
on reflection from a denser medium. 
In these diagrams, phase is indicated by 
solid and broken line segments. For 
simplicity, the angle of refraction is 
ignored. 

The high-index film produces rein- 
forcement for a specified wavelength in 
the reflected light and cancellation in 
the transmitted light. Here, the only 
phase change due to reflection from a 
denser medium occurs at the air-film 
boundary. 

The successive rays, proceeding toward 
the right, are less and less intense ; hence 
the effects of the first two reflected rays, 
and the first two transmitted rays, are 
dominating. The phase reversal of 
alternate succeeding rays modifies the 
effect of the first two rays, therefore 
neither cancellation nor reinforcement 
is complete. 

In the low-index film, there is a phase 



change in the initial reflection from the 
air-film boundary and at each reflection 
from the film-glass boundary. As a 
result, an effective film thickness equal 
to J of the wavelength here produces 
cancellation in the reflected light and 
reinforcement in the transmitted light. 

The amplitude of the reflected rays, 
for a specified angle of incidence, is 
determined by the relative indices of 
refraction of the materials on either side 
of the boundary from which reflection 
occurs. Therefore, if the relative indices 
of refraction of the air, film and glass 
were of such value that the amplitude 
of the ray initially reflected at the air- 
film boundary equalled the sum of the 
amplitudes of the other reflected rays, 
no light of the specified wavelength would 
be reflected from the film in diagram (B), 
since the rays are 1 80 out of phase. 

The maximum reflectance obtainable 
from a single film is determined by the 
index of refraction of the film. In order 
to increase the reflectance for a certain 
wavelength, multiple alternate layers of 
high- and low-index materials are ap- 
plied. The film thickness for each layer 
is controlled so that the initial reflected 
rays, of the control wavelength, from 
each film boundary are in phase when 
they leave the film. The reflected 
intensity for that wavelength is then 
equal to the square of the sum of the 
amplitudes of the reflected rays. 

A diagram representing a seven-layer 



358 



April 1953 Journal of the SMPTE Vol.60 



INCIDENT BEAM 
WAVELENGTH C/O=X 



AIR 



Fig. 2. Interference 
effects in a multi- 
layer film on glass. 



./VVAAA \ \ 




/\ /% \ \ \ \ V 

/\/\\ * \ \ \ Ni 

' \f \ \ \ \ \ \ L 

\A \ \ \ \ \ \ ' "* 



\ \ \ \ \ \ \ 




5000 eooo 

WAVELENGTH IN ANGSTROMS 

Fig. 3. Characteristics of a seven ^ X-layer dichroic. 



film and its effect on an incident light 
ray is shown in Fig. 2. In this film, each 
layer is effectively equal in thickness to 
wavelength. The initial reflected 
rays from each boundary emerge from 
the film exactly in phase. The strongest 
transmitted rays resulting from interfilm 
reflections are 180 out of phase with the 
initial transmitted ray. In addition to 
the strongest set of reflected and trans- 
mitted rays, there are many weaker rays 
resulting from interfilm reflections. 
Some of these rays are out of phase with 
the stronger rays but since they are so 



much weaker they, of course, cannot 
cancel the effect of constructive inter- 
ference in reflected light and destructive 
interference in transmitted light for 
wavelength X. 

The film design shown in Fig. 2 is a 
standard design which, when composed 
of the materials most commonly used, 
reflects over 90% of the light at the con- 
trol wavelength. The effect of this film 
on light of other wavelengths is deter- 
mined by the relative phase of the reflec- 
tive rays and of the transmitted rays as 
they leave the film. Transmission char- 



Mary Ellen Widdop: Dichroic Mirrors 



359 



(I) 




Fig. 4. Path difference of rays reflected 
from the two boundaries of a thin film. 

acteristics in the visible range for a film 
of this design, controlled for maximum 
reflectance at 6350 A at 45 angle of 
incidence, are shown in Fig. 3. There 
is no appreciable absorption: therefore, 
all light not transmitted is reflected. 
A dichroic of this type is used to reflect 
red light in the color-television camera. 
As shown in Fig. 3, the transmission 
curve shifts toward the short wavelength 
end of the spectrum when the angle of 
incidence is increased. This shift is 
due to the change of path difference with 
change of angle, making the film effec- 
tively thinner as the angle of incidence 
is increased. The path difference pro- 
duced by a film is greatest at normal 
incidence. The diagrams in Fig. 4 
represent two light rays incident on a 
thin film in air, and illustrate the fac- 
tors which determine path difference. 
If a ray strikes the film at normal inci- 
dence, the path difference between the 
ray reflected from the first surface and 
those reflected from the second is equal 
to the product of the index of refraction 
of the film and twice the film thickness. 
Diagrams 1 and 2 represent rays inci- 
dent on the film at two different angles. 
In both cases, dotted lines AC and DB 
represent successive positions of a wave- 
front. Therefore, the optical paths AD 
and CB must be equal, AD (in ah*) = 



N (CB), and the difference in optical 
path at wavefront DB, where the reflec- 
tion from the second surface leaves the 
film, is equal to N (AFC). It can be 
shown that the path difference = N 
(2d cos r), thus as the angle of incidence 
increases, the angle of refraction increases, 
the cosine of r decreases and the path 
difference decreases. 

Such a shift of light-dividing charac- 
teristics with change of angle of incidence 
is sometimes objectionable because of 
resultant shading in transmitted and re- 
flected images. Shading from one side 
of an image to the other, due to large 
field angle, can be compensated for by 
tapering the deposited film. To do this, 
the glass is placed in the evaporating 
equipment at an angle to the plane of the 
evaporator, rather than parallel to it. 
The films deposited are then wedge 
shaped instead of plane parallel. The 
resultant mirror is then positioned hi 
use so that the thickest part of the film 
receives light at the greatest angle. 

By varying the number of deposited 
layers, the thickness of each layer, the 
control wavelength and the angle of 
control, a wide variety of dichroic 
mirrors is obtainable. In Figs. 5 and 6, 
transmission characteristics are shown 
for a number of designs used to fulfill 
various light dividing or combining 
requirements. These designs consist of 
various combinations of J-, J- and f- 
wavelength film thicknesses. 

In general, designs made with layers 
thicker than J wavelength have a nar- 
rower band of reflection than ^-wave- 
length designs. The heavy solid line 
in Fig. 5 is the transmission curve for a 
mirror of the type used to reflect blue 
light in the color camera. 

To illustrate a way in which the inter- 
ference mirrors are used, a diagram of the 
optical system of an RCA Color-Tele- 
vision Camera is shown in Fig. 7. When 
light entering the camera through the 
lens system reaches the dichroic cross, 
red light is reflected from mirror J to 
totally reflecting mirror M and then to 



360 



April 1953 Journal of the SMPTE Vol. 60 




6000 

WAVELENGTH IN ANGSTROMS 
Fig. 5. Light-dividing characteristics of several dichroics. 




4000 



sooo ooo 

WAVELENGTH IN ANGSTROMS 



Fig. 6. Light-dividing characteristics of several dichroics. 



the image orthicon in the red channel. 
Blue light is reflected from mirrors I and 
K, totally reflecting mirror L and then 
to the image orthicon in the blue channel. 
Green light passes through both mirrors 
to the image orthicon in the green chan- 



nel. Each reflector efficiently transmits 
the other two primary colors. In this 
application, the dichroics are assembled 
in sandwiches with the reflecting surface 
on the inside, covered by another plate 
of the same thickness as that on which 



Mary Ellen Widdop: Dichroic Mirrors 



361 




362 



April 1953 Journal of the SMPTE Vol. 60 




5000 6000 

WAVELENGTH IN ANGSTROMS 
Fig. 8. Transmission characteristics of absorption-interference niters. 



7000 



the dichroic film is deposited. The 
light path through the glass is then equal 
for all channels. The blue reflecting 
mirrors are carefully aligned so that the 
reflecting surfaces are in one plane and 
perpendicular to the red reflecting sur- 
face. With this arrangement, astigma- 
tism introduced by the glass can be com- 
pensated for by placing two plates, E 
and F, at right angles to each other in 
front of the relay lens. The line of inter- 
section of these plates is at right angles 
to the center line of the optical system 
and to the line of intersection of the 
dichroics.* 

Materials which absorb light can some- 
times be used advantageously in inter- 
ference films. In Fig. 8, transmission 
characteristics of efficient red and green 
transmission filters are shown. These 
filters are made by depositing alternate 
layers of high- and low-index materials 
on glass, but in this case one of the mate- 
rials used has the combined properties of 



* L. T. Sachtleben, D. J. Parker, G. L. 
Alice and E. Kornstein, "Image orthicon 
color television camera optical system," 
RCA Rev., 13, No. 7: Mar. 1952. 



very high index of refraction and strong 
absorption in the blue region of the spec- 
trum. In the resultant films, blue 
light is removed by interference and 
absorption while the red and green por- 
tions of the spectrum are efficiently 
divided by interference. 

When the reflected and transmitted 
light is to be divided in a way which is 
difficult, or impossible, to achieve with 
any single film design, sets of layers may 
be combined on one plate. For the 
final characteristics of a film to be the 
product of two different designs, the 
sets of layers must be separated from 
each other sufficiently to prevent much 
of the interference between the sets. The 
curve shown by the broken line in Fig. 9 
is that of a plate coated with two sets of 
layers, one having peak reflectance for 
blue light and the other having peak 
reflectance for red light. In this case, 
the sets of layers were separated from 
each other by a thick evaporated layer of 
material. The solid-line curve in Fig. 9 
is the resultant characteristic of two 
similar sets of layers deposited on opposite 
surfaces of a plate. Here the elimina- 
tion of interference between the sets of 



Mary Ellen Widdop: Dichroic Mirrors 



363 




5000 (000 

WAVELENGTH IN ANGSTROMS 



Fig. 9. Light-dividing characteristics obtained by depositing 
two sets of dichroic films. 



100 




( TRANSMISSION 
OF GLASS) 



364 



WAVELENGTH 
TRANSMISSION CHARACTERISTICS OF A HEAT TRANSMITTING MIRROR 

Fig. 10. Transmission characteristics of a heat-transmitting mirror. 
April 1953 Journal of the SMPTE VoL 60 



layers is more nearly complete. Highly 
reflecting films can be applied to both 
surfaces of a glass plate advantageously 
in many cases; however, this scheme 
cannot be used in any case where double 
images are undesirable. 

Transmission characteristics for an 
interference mirror which reflects effi- 
ciently through the visible spectrum and 
efficiently transmits infrared radiation 
are shown in Fig. 10.* In this case, 
several sets of layers controlled for maxi- 
mum reflection at different positions in 
the visible range, are deposited on one 
surface of a plate. In this case, no 
attempt is made to prevent interference 
between the sets of layers. Interference 
between the sets of layers improves the 
efficiency of the film. 

Interference mirrors can be made 
which reflect all wavelengths of the 
visible spectrum with approximately 
equal intensity. Such mirrors can be 
made to reflect as much as 45% of the 
incident light. A relatively thin layer 
of low-index material is first deposited 
on the glass surface, followed by a thicker 
layer of a material with a high index of 
refraction. The interference effects of 
such a film design cause all wavelengths 
of the visible spectrum to be reflected 
with approximately equal intensity be- 
cause of the relative thickness of the 
component layers for each wavelength. 
The reflectivity of such a film is plotted 
with respect to the film thickness in Fig. 
11. 

The low-index film is slightly less than 
J-wavelength thick for blue light. The 
high-index material is deposited until a 
reflection maximum is passed and the 
reflection drops to about 85% of the 
maximum. Since the low-index layer 
is not J-wavelength thick at the point of 
maximum reflectance, the interfering 
rays are not exactly in phase. This 
maximum is, therefore, not as high as it 




* G. L. Dimmick and M. E. Widdop, 
"Heat transmitting mirror," Jour. SMPTE, 
58: Jan. 1952. 



*A ix Jx 

FILM THICKNESS IN WAVELENGTHS 

Fig. 11. Design of an achromatic film. 

would have been if the low-index film 
had been J-wavelength thick. 

For longer-wavelength green light, the 
effective thickness of the low-index film 
is less than for the blue light. The maxi- 
mum intensity obtainable for this wave- 
length is, therefore, less than it was for 
blue. For these film thicknesses then, 
the reflected intensity for green light has 
just passed its maximum point, and is 
approximately equal to the intensity 
for the blue. And for the red light, 
having the longest wavelength, the inter- 
ferring rays are at a point of maximum 
reinforcement, but the resultant intensity 
is equal to that of the green and blue. 
The transmission characteristics of a 
mirror of this type are shown in Fig. 12. 

Discussion 

A. V. Loughren, Chairman of the Session 
(Hazeltine Corp.} : The importance of these 
dichroic mirrors in the development work, 
which has been going on recently both at 
RCA and many other places in color tele- 
vision, is something to which I can certainly 
testify personally. Our own developments 
at Hazeltine came along far faster than they 
would have otherwise because it was pos- 
sible to make arrangements to get some of 
the dichroic mirrors designed by Miss Wid- 
dop and use them in our own work. I have 
no doubt myself as to their importance. 

L. L. Ryder (Paramount Pictures Corp.) : Is 
any work being done in conjunction with 
these mirrors for high-intensity light levels, 
for instance where they're used in conjunc- 
tion with arcs and similar light sources? 



Mary Ellen Widdop: Dichroic Mirrors 



3*5 



iMITTANCE 
S I 









20 








cr eo 

1- 


" ' 








40 

. 

60 








c 








80 

100 
00 



40 








00 5000 6000 7( 



WAVELENGTH IN ANGSTROMS 
Fig. 12. Characteristics of an achromatic light divider. 



Miss Widdop: Well, the heat-transmitting 
mirror which was the last design shown, 
was developed primarily with that idea in 
mind, that it would efficiently reflect the 
visible light and transmit heat, but it never 
has actually been applied to that purpose. 

Mr. Ryder: The real question is will the 
coating material stand up against these 
extreme heats of the arc? 

Miss Widdop: We don't feel that the films 
could be used as front surface mirrors. If 
they were on the back of the mirror or if 
they were protected, the heat would not 
bother them, as far as our experience has 
shown. We haven't actually had such a 
mirror in use for a long period of time in a 
high-intensity arc, but we have made tests 
with it when it has been in a number of 
hours. I realize this wouldn't compare with 
the number o hours a mirror would be 
used in practice. The difficulty with put- 
ting such a coating on the back surface of a 
curved mirror is that it has to be very uni- 
form or a pronounced color will be re- 
flected. But the coating could be put on 
the inside surface and another glass put in 
front to protect it. Or a projector could be 
redesigned so that the beam is reflected at 
an angle from a flat mirror and then 
through the film. 



Mr. Ryder: A question pertaining to the 
resolving power of lenses or lens systems 
which have the dichroic mirrors: How do 
lens systems using dichroic mirrors com- 
pare with lens systems with no reflection 
mirror? My interest is lens systems for 
motion picture or television cameras. 

Miss Widdop: I'm not sure that I can 
answer that question. The glass on which 
the films are deposited has to be very flat, 
and compensating plates are used to correct 
for astigmatism introduced by the thickness 
of the glass and the angle of the mirrors. 
The films themselves do not introduce any 
distortion. 

Ralph Heacock (RCA Victor Div., Camden, 
N.J.) : In answer to Mr. Ryder's comment 
on heat-reflecting mirrors from a practical 
field viewpoint, we made use of some heat- 
reflecting mirrors and found that they 
would not satisfactorily stand up in the field 
under high-intensity arc light projection. 
More recently, however, we find with the 
heat-reflecting, inclined mirror, that if a 
draft of cool air is directed over the reflec- 
tor, it operates satisfactorily over long 
periods of time. The maximum amperage 
that they have used them under, as far as I 
know, has been about 105 amp with a 10- 
mm positive carbon. 



366 



April 1953 Journal of the SMPTE Vol. 60 



ABSTRACT 



Television Recording 

By W. D. KEMP 



Various means of recording vision-frequency signals are considered. For 
high-definition television, photographic recording appears to be the only 
practicable method. The film may be moved at constant velocity or inter- 
mittently, and some known systems are classified under these headings and 
described with their advantages and disadvantages, particular attention being 
given to the two methods now in use by the B.B.C. Various factors affecting 
the performance of television-recording cameras are then discussed, such 
as picture joins, interlacing, loss of definition on movement, optical efficiency, 
film stability, emulsion pile-up, and spot-position modulation. A suitable 
arrangement of the vision-frequency equipment is dealt with, and the neces- 
sary electrical corrections are described. The factors governing the choice 
of a cathode-ray tube for recording purposes are discussed, while some pos- 
sible improvements are indicated. The various transfer characteristics con- 
cerned have been measured. Contrast correction is discussed, together with 
possible ways of improving the overall performance, such as high-gamma 
recording. The conditions for making satisfactory positive prints direct 
from negative television pictures are stated. Some conclusions based on 
operating experience of continuous and intermittent systems are then given. 



(1) INTRODUCTION 

Still photographs of high-definition In Britain, initial development was 

pictures are readily made, but photo- concentrated on 35mm equipment, but a 
graphing such pictures on motion pic- 16mm system has now been developed. 

ture film in a form suitable for retrans- The B ' B - C - consid * r the devel P men < f 
,.fr , high-quality recording systems on both 

mission presents some difficulty. A num- / c . , f 

16mm and 35mm equipment to be of 

her of methods have, however, been prime i mport ance owing to the possi- 
suggested and some of these are being bility of interchange of programmes with 
used successfully. other countries. 

Abstract by Pierre Mertz of a paper first N. 5, England. This abstract is published 

presented on November 26, 1951, and in through the cooperation of The Institution 

revised form on March 11, 1952, at the of Electrical Engineers, Savoy Place, 

Convention on the British Contribution to London W.C. 2. 

Television, April 28 - May 3, 1952, by This abstract is comprised of about 75% 

W. D. Kemp, High-Definition Films of the original paper. All illustrations 

Limited, 98 Highbury New Park, London except Fig. 8 have been retained. 

April 1953 Journal of the SMPTE Vol. 60 367 



*(3) SYSTEMS OF TELEVISION RECORDING 



These can be divided into two groups : 
those in which the film moves continu- 
ously, and those in which the film moves 
intermittently as in normal motion pic- 
ture practice. 

(3.1) Continuous-Motion Types 

To record interlaced television by this 
method, the geometrical displacement 
error over the extremes of movement 
must be less than one picture element, in 
both line and frame directions. 

(3.7.7) Cameras Employing Picture Move- 
ment: In this system, the television picture 
on the cathode-ray-tube face is moved 
physically in such a direction that it re- 
mains stationary relative to the film, 
which is moving continuously at 25 
frames/sec. The movement of the pic- 
ture is achieved by electrical deflection of 
the raster. The high degree of precision 
to be attained has precluded the develop- 
ment of a practical television-recording 
system using this principle. 

(3.7.2) Cameras Employing a Picture of 
Special Aspect Ratio and Double Optical 
System: This is the normal method of 
transmitting film by means of the flying- 
spot principle, but used in reverse. The 
film moves continuously at 25 frames/sec 
and two images of the television picture, 
which is of approximately twice the nor- 
mal aspect ratio, i.e. 8:3, are focused on 
the film, separated by half a film-frame 
pitch. The film moves half a frame pitch 
during the recording of the first field and 
this movement restores the aspect ratio 
on the film to the normal value. During 
this period the second image is obscured 
by a shutter. The shutter then obscures 
the first image and allows the second 
image to be exposed to the film, the lines 



* The section numbers in this abstract are 
not consecutive. The author's original 
section numbers have been retained for 
the sake of consistency in reference to the 
original paper. Ed. 



falling between those of the first field, 
since the film has now moved to the re- 
quired position. 

(3.7.3) Shutter less Cameras Employing 
Optical Compensation: There are numerous 
designs utilizing this principle. Most of 
them were intended for cinema projection 
where it was hoped to increase the screen 
brightness and reduce the flicker con- 
tent, owing to the elimination of the 
shutter intervals. 

Despite the theoretical advantages of 
continuous-motion projectors, compared 
with those having intermittent motion, 
they have never proved popular in the 
motion picture industry, probably be- 
cause of the complexity of the various 
designs. The Mechau projector, how- 
ever, enjoyed moderate popularity in 
Germany and an improved version of 
this machine has been converted into a 
recording camera for B.B.C. use. Apart 
from the Mechau projector, only one 
other type of optical-compensation pro- 
jector has been used to any extent; this 
is the type employing a thick polygonal 
prism. 

The system is extremely simple and 
involves a minimum of moving parts; 
for this reason it is the basis of most high- 
speed cameras, exposure rates of up to 
3000 frames/sec being attainable. The 
difficulty for adaptation to television lies 
in the accuracy necessary in manufactur- 
ing the prism, the large size of the prism 
required for high definition and the prob- 
lem of adjustment for film shrinkage. A 
telecine projector working on this sys- 
tem has, however, been built in the 
United States, and at least one home pro- 
jector of similar design has been marketed 
in Germany. 

The Latest B.B.C. 35mm Continuous- 
Motion Recording System: The first B.B.C. 
35mm recording suite went into service in 
October 1949. The design utilized the 



368 



April 1953 Journal of the SMPTE Vol. 60 



FILM MAGAZINE 



CATHODE-RAY-TUBE SCREEN 




FLOOR LEVEL 



Fig. 1. Schematic of the latest B.B.C. 35mm continuous-motion 
recording system. 



Mechau projector with the original 
Mechau film-transport mechanism. 3 

Experiments carried out on this equip- 
ment showed that the Mechau mirror 
drum was fully capable of providing cor- 
rect optical compensation. The film- 
transport mechanism, however, did not 
attain the same high standard; it was 
possible to detect velocity modulation at 
the sprocket-hole frequency of 100 cycles 
and at other higher frequencies. This 
prevented accurate interlacing on the 
film. 

It was therefore decided to redesign the 
equipment completely, retaining only the 
mirror and its associated optical system, 
and three machines of this new type are 
shortly to be installed at the Lime Grove 
television studios. 

Figure 1 is a schematic arrangement of 
the equipment. The optical system con- 
sists of the collimating lens LI, the fixed 
mirror M, the mirrors mounted on the 
mirror drum, Ml, M2...M8, and the 
main lens L2. At the particular posi- 
tion of the mirror drum shown, Ml is pro- 
ducing an image of a point P (on the 
cathode-ray tube) at P'. As the drum 



rotates, P' moves down with the film. 
When P' has approached the end of its 
travel, the cone of light projected by M 
on to Ml starts to fall on M2, and an 
image is produced at P''. The distance 
between P' and P" is equal to one film- 
frame pitch. As the image P' decreases 
in intensity, the image P" increases in 
intensity, since the luminous flux is 
shared between P' and P". 

The exposure sequence is shown in 
Fig. 2, where (a) represents the television- 
frame waveform and (b) the exposure 
produced on consecutive film frames by 
consecutive mirrors in the original 
Alexandra Palace equipment. It will 
be seen that, although the mean exposure 
period was 0.04 sec, some portion of 
exposure was received over a period 
greater than two scans. This resulted 
in a loss of definition on moving objects 
in the picture, owing to individual film 
frames being exposed partially to preced- 
ing and following television frames. 

When the diameter of the cone of rays 
falling on the mirrors is smaller than the 
gaps between the mirrors, the exposure 
sequence is as shown in Fig. 2(c). There 



W. D. Kemp: Television Recording Abstract 



369 




r ' ~~V 

W 



M2 



Fig 2. Exposure sequence for the B.B.C. continuous-motion recording system. 

(a) Television-frame waveform (Fl, F2, etc. frame period. PI, P2, etc. complete 
picture period). 

(b) Exposure produced on consecutive film frames by consecutive mirrors in the original 
B.B.G. equipment (Ml, M2, etc. exposure period). 

(c) Exposure sequence when the cone diameter of the rays is less than the gaps between 
the mirrors. 



is now a period when no light falls on the 
film, and it is therefore necessary to so 
phase the mirror drum that the mirror 
change-over period occurs during the 
television frame-suppression period. A 
special motor-drive unit provides a very 
accurate phase lock to the television 
frame-synchronizing signals. The phase 
position is indicated by stroboscopic 
means. 

The film-traction mechanism of the 
original Mechau equipment employed a 
four-picture sprocket to pull the film 
through the curved gate. The new film 
path is shown in Fig. 3. The sprocket 
SI withdraws film from the magazine, 
and, after a free loop, the film passes to a 
smooth drum Dl, which is mounted on 
the shaft of an eddy-current brake. After 
travelling through the curved gate G, the 
film passes round a second smooth drum 
D2, which has a flywheel of large inertia 
coupled to it. A small roller R serves to 
wrap the film round this drum. The 
film then passes via a sprocket S2 to the 
sound-recording drum D3 and thence 
back to SI and to the magazine. S2 
provides the motive force to pull the 



film through the picture gate. The 
framing can be varied by rotating R 
about the centre of D2. 

When static, there is no tension in the 
picture gate. The running tension is 
supplied entirely by the braking action 
of Dl. The inertia of D2 is so consider- 
able that it would take a long time for the 
film to accelerate it from rest, and 
advantage is taken of the lack of gate 
tension under static conditions by arrang- 
ing that this flywheel is kept rotating at 
full speed by means of a separate electric 
motor. Initially D2 withdraws a small 
amount of film from the gate, but, as the 
film is then slack round the drum, no 
further film can be withdrawn until the 
main motor of the camera, which drives 
SI and S2, is started. Additional con- 
tacts on the starting relay then actuate a 
magnetic clutch which disconnects the 
electric motor from D2 so that D2 is 
driven by the film alone. 

The film is held against the gate sur- 
face by two curved runners independently 
spring-loaded. The spring-loading is 
kept to an absolute minimum to avoid 
gate friction, which is liable to vary and 



370 



April 1953 Journal of the SMPTE Vol, 60 



so produce velocity modulation. The 
complete gate assembly can be withdrawn 
for cleaning. 

It is hoped that the new film-traction 
mechanism will produce velocity sta- 
bility adequate to obtain correct inter- 
lacing on the film. 

(3.2) Intermittent-Motion Types 

(3.2.7) Fast Pull-Down in Frame-Sup- 
pression Period: This is the simplest and 
most attractive method, but it involves a 
pull-down and registration period of 
about 12 on present B.B.C. standards, 
and, so far as is known, this figure has 
never been achieved. 

(3.2.2) Cameras Employing Two Films: 
The proposal has been made to use a 
double optical system in conjunction 
with two gates which would have film 
movements 180 out of phase. The 
first television picture would be recorded 
in gate 1 , and then, while the film in this 
gate was being advanced, the next tele- 
vision frame would be recorded in gate 2. 
Each film, after processing, would have 
half the television pictures on it, and a 
step printing process would be required 
to produce a 25-frames/sec film. 

The principal technical disadvantage 
of any two-film system is the difficulty of 
ensuring identical processing. Even 
when the two films are joined together 
and passed through the same processing 
machine, variations along the length are 
difficult to eliminate with the required 
degree of accuracy. 

Obviously, the cost of any two-film 
system is much higher than that of a 
single-film system. The time required 
to produce a finished print is also con- 
siderably greater. 

(3.2.3) Cameras Involving Memory Sys- 
tems: If it were possible to delay every 
other television field by 0.02 sec, then 
two consecutive fields could be recorded 
simultaneously on the film and so the 
period of one field would be available for 
pull-down. 




Fig. 3. Film-traction mechanism for the 
latest B.B.C. 35mm continuous-motion 
recording system. (Ml, M2, etc. 
exposure period.) 



A suggestion 4 involving the use of a 
fluorescent screen for retaining the first 
field has been made. The objection to 
this method is the characteristic re- 
quired of the fluorescent material. To 
avoid frame shading it would have to 
possess a decay time long enough to allow 
negligible falling-off in brightness during 
the time that information was being re- 
corded. At the same time the brightness 
should have fallen to a negligible value 
after the next exposure period. The 
exponential decay characteristic associ- 
ated with normal phosphors appears 
unsuitable for this purpose. 

There has not yet been a practical de- 
velopment of a camera working on this 
principle. 

(3.2.4) Use of Image Converter for Change 
of Standards: It has been suggested 5 that 
the television signals to be recorded 
should be applied to an image converter 
which would change the scanning stand- 
ards and give a greater frame-suppression 
period to allow an adequate pull-down 
period of, say, 30 or more. 

An image converter would enable the 
incoming interlaced picture to be changed 



W. D. Kemp: Television Recording Abstract 



371 



CATHODE-PAY 
TUBE 



ILM 




OP APERTURE 
OTTOM APERTURE 
LAW MCCHANISM 



Fig. 4. Schematic of the camera for the Kemp-Duddtngton system. 




OPEN CLOSED OPEN CLOSED OPEN CLOSED 



CYCLIC PERIOD 1/24 SEC 



UPPER APERTURE 
OPEN 





Of 


EN 


CUOttTD 


O 


EN 

; 


CLOSED 


OSEC 


WA 


CYCLIC 


PERIOD 3/5 








(a) 



(b) 



(c) 



(d) 



(e) 



(f) 



Fig. 5. Waveforms for 
intermittent-motion sys- 
tems. 

(a) American television 
waveform (PI, P2, 
etc. complete picture 
period). 

(b) Shutter sequence at 
present used in 
America. 

(c) Alternative American 
shutter sequence using 
double optical system. 

(d) B.B.C. television wave- 
form. 

(e) Exposure sequence for 
16f method. 

(f) Exposure sequence for 
Kemp - Duddington 
system. 



372 



April 1953 Journal of the SMPTE Vol.60 



into a sequential picture, which might 
ease some of the recording problems. 

The principal objection to the method 
is the additional complexity and the dis- 
tinct possibility of loss of quality in the 
conversion process. 

(3.2.5} American Method: Although this 
is not applicable to British standards, 
since only 24 of the 30 television pic- 
tures/sec used in America 6 are recorded, 
it is interesting to compare the exposure 
cycle with those of the British methods 
described in Sec. 3.2.7 and 3.2.8, which 
are suitable for 25-pictures/sec television 
recorded at 25 film frames/sec. 

Figure 5 (a) shows the American tele- 
vision waveform diagrammatically, while 
Fig. 5(b) shows the shutter sequence at 
present used in America. It will be seen 
that every alternate field is joined in the 
picture area. With the relative phasing 
of camera and television signals as shown, 
this join occurs in the centre of the pic- 
ture ; with other conditions of phasing, a 
join in the picture area occurs on each 
field. Picture joins are dealt with in 
Sec. 4.1. 

(3.2.6) Alternative Exposure Sequence for 
American Standards: This uses a double 
optical system and double-gate aperture, 
the images being one above the other. 
The shutter sequence is shown in Fig. 
5(c), and a schematic of the camera in 
Fig. 4. Television picture 1 is recorded 
in the bottom aperture, then television 
picture 2 in the top aperture. One field 
of television picture 3 is then missed 
while the film is advanced two frames 
and the cycle is repeated. The advan- 
tage of this arrangement is that there is no 
picture join. 

The method is of particular interest, 
since the camera developed to record 
British standards on the Kemp-Dudding- 
ton system could be adapted to record 
American standards on the above se- 
quence. A pull-down period of 72 
would be required, compared with the 
90 on British standards, but it is felt 



that this could be achieved without 
radical modification. 

(3.2.7} 76% Method*: This uses the time 
of every third scan for pull-down, thus 
providing 16f film frames for every 25 
television pictures. A step printing 
operation to print every other film frame 
twice is required to obtain 25 frames/sec 
on the film. The reproduction of move- 
ment is, of course, not so good as in a 
25-frames/sec method but should not be 
worse than that associated with the step 
printing of 1 6-frames/sec motion pic- 
tures to 24 frames/sec. The exposure 
sequence is shown in Fig. 5(e), Fig. 5(d) 
being the B.B.C. television waveform. 

(3.2.8) Kemp-Duddington Method 6 : This 
is the system which the B.B.C decided to 
develop for 16mm recording. 9 A double 
optical system and two-frame pull-down 
is required. The exposure sequence is 
shown in Fig. 5(f), and the camera 
schematic is the same as that of Fig. 4. 

The first and second scans of television 
picture 1 are recorded by image 1 in the 
bottom aperture, and the second scan of 
picture 1, together with the first scan of 
picture 2, is recorded on the second film 
frame in the top aperture. The second 
scan of television picture 2 is not re- 
corded, the film being advanced by two 
film frames in this period. The cycle is 
then repeated. With the phasing shown, 
no picture join is necessary. To ensure 
that the correct phasing is maintained, 
the camera is phase-locked to the tele- 
vision-frame pulses and the relative phase 
angle is indicated by stroboscopic means. 

It will be seen that some 90 of the 
cycle are available for film pull-down, 
and in the design of the camera an 
accelerated triangular cam-movement 
was used giving a pull-down of 60, with 
15 allowed for registration. This was 
done by means of register pins fixed to the 
aperture plate. 

Optical registration of the images is 
necessary, and means are provided for 



W. D. Kemp: Television Recording Abstract 



373 



achieving this to a high degree of ac- 
curacy. It is also necessary, although not 
difficult, to balance very exactly the 
exposure produced in each gate aperture 
so that 12.5-cycle flicker is avoided. 
The reproduction of movement is 



theoretically worse than with an ideal 
25-frame television-recording system. 
However, tests have shown that, in fact, 
movement is reproduced very satisfac- 
torily and compares with normal motion 
pictures. 




(a; 



(c) 




(f) (9) 

Fig. 6. Diagram illustrating the effect of picture joins. 

(a), (b) and (c) show the effect of the shutter opening. 

(d) shows the lines recorded during the last scan. 

(e) shows the picture join obtained. 

(f), (g) and (h) illustrate the fading-shutter method. 



(4) SOME FACTORS AFFECTING THE PERFORMANCE OF TELEVISION- 
RECORDING CAMERAS 



(4.1) Picture Joins 

In the American system of television 
recording and in the 16| or Kemp- 
Duddington systems, when used in a 
general phase relationship, picture joins 
occur in the picture area. With a cath- 
ode-ray-tube phosphor of negligible 
afterglow and using a focal-plane shutter, 
the sequence of operation is shown hi 



Fig. 6. As the shutter opens after the 
pull-down period, lines should be re- 
corded of increasing length owing to the 
diminution of the occulting effect of the 
shutter edge. This is shown in Figs. 
6 (a), 6(b) and 6(c). For the next scan 
the shutter is open, and in the last part 
of the cycle the lines should be recorded 
of decreasing length and in such a posi- 
tion as exactly to join with those recorded 



374 



April 1953 Journal of the SMPTE Vol.60 



one picture period previously. Figure 
6(d) shows the lines recorded during the 
last scan, and Fig. 6(e) the picture join 
obtained. The effect of afterglow is to 
blur the edges of the join. 

It will be appreciated that, to obtain 
such a perfect join, the shutter angle and 
velocity stability of the shutter have to be 
substantially perfect. In the Eastman 
Kodak television-recording camera for 
American standards, the shutter is driven 
by a separate synchronous motor and the 
shutter edges are stoned down on test to 
produce an exact shutter angle. Even 
so, the picture join could be detected on 
some early American recordings. 10 

Figures 6(f), 6(g) and 6(h) show a pro- 
posed solution to this problem by the use 
of a fading shutter. 8 In this method, the 
shutter is caused to fade in the lines 
gradually, instead of occulting them. One 
picture period later the lines are faded 
out, the fade-in zone of the picture over- 
laying the fade-out zone. 

An alternative method of achieving a 
2food picture join is to employ blanking 
of the recording monitor. 11 In this case, 
the cathode-ray tube is normally biased 
Deyond beam cut-off. When the film is 
n the gate and ready for exposure, the 
camera sends a trigger pulse which 
aises the cathode-ray-tube bias to the 
correct figure for exactly one picture 
Deriod (the correct number of lines being 
counted electrically), after which the 
cathode-ray tube is again biased back 
and the film is pulled down. There is 
no shutter on the camera. With this 
system, lines recorded just prior to film 
pull-down are not exposed to the after- 
glow and tend to be under-exposed. As 
the joint occurs only on alternate frames, 
the variation in density tends to cause 1 2- 
cycle flicker. The effect can be reduced 
by reducing the pull-down time and 
phasing so as to allow additional expo- 
sure time after each television frame. 
The picture joins produced by this sys- 
tem can be undetectable unless move- 
ment has occurred during the picture 



period when there is a discontinuity at the 
join. 

(4.2) Interlacing 

In order to record an interlaced pic- 
ture by means of a continuous-motion 
process, it is necessary that the film and 
scanning spot should move in such a 
fashion that after an interval of 0.04 sec 
the scanning spot is in the correct posi- 
tion relative to the film within a small 
percentage of the line pitch. Even with 
a 405-line picture, this implies a high 
degree of velocity stability for the film 
movement. The difficulties of eliminat- 
ing sprocket-hole velocity modulation 
(or "flutter") and slow-speed velocity 
modulation (or "wow") are well known 
to motion picture sound-recording engi- 
neers, but in their case the integrated 
displacement error over any period of 
time is relatively unimportant, providing 
the period is long. 

Much work on the stabilization of film 
velocity for television purposes has been 
carried out in connection with flying-spot 
telecine machines, and it is a tribute to 
the engineers concerned that substan- 
tially 50:50 interlacing with a picture 
float of less than a line pitch has been 
obtained. It should be noted that it is 
the error integrated over the 0.04-sec 
period that is important, and advantage 
has been taken of this in selecting the 
rotation speed of the stabilizing flywheel 
used in the latest B.B.C. continuous- 
motion recorder already described, where 
sub-harmonics of 25 cycles were carefully 
avoided. 

Owing to the nature of the film base, 
the linear dimensions of motion picture 
film are variable with its age and the 
humidity and temperature of the storage 
conditions. With triacetate base, used 
for noninflammable film, the sprocket- 
hole pitch of raw stock is usually within 
0.1% of the nominal dimensions, a 
shrinkage having occurred since perfora- 
tion. This shrinkage may rise to 0.2% 
or more if some months are allowed to 
elapse between perforation and use. 12 



W. D. Kemp: Television Recording Abstract 



375 



ORIGINAL 
TELEVISION SCANS 



RE-TRANSMITTED 
TELEVISION SCANS 



SCAN 2 




FILM FRAMES 






(b) 



(c) 



Fig. 7. Diagram illustrating the use of spot-position modulation 
to eliminate the line structure on recorded film. 



Nitrate base, which is inflammable, has 
the same order of shrinkage initially, but 
after longer periods of time tends to 
shrink more than the triacetate base. 13 

In the systems described in Sec. 3.1.3, 
the linear displacement between the two 
images must depend on film shrinkage, 
and in the systems described in Sec. 3.1.1 
and 3.1.2, the displacement of the image 
during the cycle will be required to vary 
to the same degree. In the present 
B.B.C. continuous-motion system, the 
displacement is variable and is set up 
using a sample of the film stock which will 
be used for recording. The variations 
from roll to roll are then small compared 
to the television line pitch, which is 
approximately 0.25% of the picture 
height. This is true, however, only 
when very careful check is kept on the 
stock used, and the dependence of inter- 
lacing on film dimensions must be 
reckoned to be a disadvantage of con- 
tinuous-motion recording. 



Intermittent-recording methods do not 
suffer from the above difficulties since the 
film is stationary during the exposure 
period and tolerances in perforation 
pitch are taken up in the film-advance- 
ment mechanism. This is the main rea- 
son for the choice of intermittent motion 
for B.B.C. 16mm recording, where 
difficulties are increased by the higher 
magnification. 

A further problem of recording inter- 
laced television is the possibility of physi- 
cal movement between the two scans due 
to vibration. Vibration frequencies of 
25 cycles (or 25-cycle harmonics) are 
particularly objectionable, as already 
explained, and long-focus lenses increase 
the difficulty, since a small angular 
vibration will then produce a bigger 
linear displacement. 

The difficulties of obtaining interlacing 
on the film are illustrated by the very 
high percentage of still photographs 



376 



April 1953 Journal of the SMPTE Vol.60 



taken of interlaced television pictures 
which show 188f-line structure. 

(4.3) Loss of Definition on Movement 

A television programme derived from a 
live scene provides 50 images/sec of the 
subject, but two of these are recorded on 
each film frame, since the film runs at 25 
frames/sec. When the film is retrans- 
mitted, it is unlikely that the two scans 
will be reproduced in the correct order. 
In practice, owing to imperfect frame- 
deflection linearity, portions of both 
scans are transmitted in different zones 
of the picture height. The use of spot- 
position modulation (see Sec. 4.7) to 
eliminate the line structure on the re- 
corded film makes it even more difficult 
to separate the scans. This is illustrated 
in Fig. 7. Four consecutive television 
scans of a subject consisting of a vertical 
line moving horizontally at a rapid rate 
are shown in column (a). Television 
scans 1 and 2 are recorded on film frame 
1, and scans 3 and 4 on film frame 2, 
shown in column (b). Column (c) 
represents the retransmission of the re- 
cording when spot modulation is used. 
It will be seen that a doubling of the 
image has occurred. When the rate of 
movement is less than that shown an 
irregularity on vertical edges is produced. 

The doubling of images on fast move- 
ment is more objectionable to the eye 
than the blurring produced by the nor- 
mal motion picture camera. With tele- 
vision cameras involving storage, how- 
ever, the doubling is not so apparent 
since the original images do not have dis- 
tinct edges. 

(4.4) Optical Efficiency and Resolution 

Optically compensated continuous- 
motion systems require a movement of 
the image during exposure, and this 
necessitates a greater coverage for the 
lens. To keep the image in focus during 
its travel, the film is usually moved 
round a curved path, the lens being de- 
signed to have a spherically curved field, 
of radius equal to that of the gate. A 



second cylindrical lens is usually neces- 
sary to flatten the field in a plane trans- 
verse to the direction of movement. 

The above considerations do not make 
for high resolution, particularly as the 
use of mirrors often necessitates addi- 
tional lenses to collimate the light and so 
prevent double images. It is difficult to 
obtain a high aperture with such sys- 
tems. The B.B.C. continuous-motion 
system described gives an effective aper- 
ture of about //3.O. 

The Kemp-Duddington 16mm camera 
has the merit of using the same lens for 
both images. The effective aperture is 
about //3.3. The resolution is quite 
satisfactory; it is possible to photograph 
an optical test pattern of 1000 picture 
elements per picture height with good 
depth of modulation. 

(4.5) Film Stability 

The defect known as picture float has 
been referred to in Sec. 4.2, and a con- 
tinuous-motion film-traction mechanism 
has been described (see Sec. 3.1.3). It 
has been found possible to reduce this to 
small proportions, although it is always 
present to some extent in continuous- 
motion recordings. More rapid varia- 
tions of picture position, known as 
"jump" or "bounce," are due to lack of 
registration of one or more film frames 
with the sprocket holes. Picture jump 
can occur, for instance, in the mirror- 
drum type of recorder when one or more 
mirrors are incorrectly set, or in inter- 
mittent recorders when each film frame 
is not properly registered. 

Correct registration is of paramount 
importance, and it is normal film prac- 
tice to have separate register pins apart 
from the claw mechanism used to ad- 
vance the film. The Kemp-Dudding- 
ton camera follows this procedure, being 
provided with a lifting gate and registra- 
tion pins integral with the camera aper- 
ture. Very good film stability is ob- 
tained by this arrangement. In cameras 
where the pull-down is fast, such as the 
Eastman Kodak for American standards 



W. D. Kemp: Television Recording Abstract 



377 



(pull-down period 72), separate regis- 
tration pins are not provided and the 
elimination of jump is a major difficulty. 

The position of the film in the focus 
plane is also liable to vary, particularly 
in continuous-motion systems, and this 
gives rise to focus float. The effect is 
most serious where long-focus lenses are 
used, owing to the small depth of focus 
in the image plane. It is found neces- 
sary to focus the B.B.C. continuous- 
motion camera by actually exposing film 
at various focus settings. This is not 
required in the intermittent Kemp- 
Duddington camera, where the film plane 
is absolutely fixed by the shuttle gate and 
aperture plate. 

Emulsion pile-up is the main cause of 
focus variations over long periods of time. 

(4.6) Emulsion Pile-up 

The emulsion on raw stock is quite soft, 
and during the exposure of a 1000-ft roll 
of film, small particles are apt to stick to 
any portion of the gate surface which is 
rough, or where there is local pressure. 
Friction and local heating follow, and 
the particles build up to solid flakes which 
hold the emulsion surface away from the 
image plane. Great trouble from this 
was experienced in the early days of 
B.B.C. recording. It was completely 
overcome by obtaining hard chrome- 
plating on all gate surfaces and oiling the 
film. The oil used was pure sperm oil, 
which was applied by means of two small 
felt pads spring-loaded so that they made 
contact with the film adjacent to the 
sprocket holes on the emulsion side, prior 
to the film entering the gate; the picture 
area on the film was not oiled. A single 
drop of oil applied to the felt pads when 
each reel of film was loaded was found to 
be sufficient. Sperm oil forms an emul- 
sion with film developer, and, provided 
the quantities of oil used are very small, 
no ill effects result. 

Intermittent-motion cameras are par- 
ticularly susceptible to emulsion pile-up, 
since there is time for the emulsion to 
stick to the gate surface and the high 



initial acceleration of the film then de- jj 
taches particles. In America use has 
been made of nylon gates which can be ij 
finished with a high polish. Another J 
method used is pneumatic holding of the J 
film in the gate aperture so that the nor- | 
mal pressure pad can be dispensed with. 11 j 
The shuttle gate used in the Kemp- 
Duddington camera is good in this re- 
spect as there is no pressure on the film ; 
while it is being moved, the film being 
quite loose in the gate slide. 

(4.7) Spot-Position Modulation 

Providing the resolution of the film 
stock used is adequate, a correctly inter- 
laced television recording contains 377 j 
lines in the picture height. When this is 
retransmitted on the same standards, the 
second scanning raster will not neces- 
sarily fall over the recorded raster, as 
has been explained in Sec. 4.2. More- 
over, for transcription purposes, the i 
second scanning raster may have different 
standards, so that registration of the t\vo 
rasters is impossible. 

The effect of this lack of registration is i 
to produce patterns of light and shade 
over the transmitted picture, the dark 
portions corresponding to the places 
where the scanning spot is traversing the 
unexposed spaces between the recorded 
scanning lines. The patterns may 
change in a random manner owing to 
the random nature of the errors, or may 
be constant if the accuracy of scanning is 
sufficient. The patterns can be com- 
pletely eliminated by eliminating the 
line structure of the image by a technique 
known as spot wobbling. 14 

This consists of applying a small high- 
frequency sinusoidal deflection to the 
electron beam of the recording cathode- 
ray tube in order to increase the effective 
height of the scanning spot without in- 
creasing its width. The sinusoidal na- 
ture of the deflection tends to produce 
the effect of a double scanning line be- 
cause of the low velocity at the maxi- 
mum and minimum limits of deflection 
and the high velocity in the mean posi- 



378 



April 1953 Journal of the SMPTE Vol. 60 



tion. When the amplitude of the deflec- 
tion is adjusted to a critical value, the 
apparent number of lines in the picture 
is thus doubled. The size of the scan- 
ning spot and limiting resolution of the 
film emulsion and optical system then 
prevent line structure being recorded. 
In America the spot-wobbling tech- 



nique is not normally applied because 
the resolving power of the 16mm film, 
mostly employed, is not adequate to 
produce a well-defined line structure 
with 479 lines in the picture height. 
The more extensive use of 35mm film in 
America would almost certainly render 
spot-position modulation necessary. 



(5) THE VISION-FREQUENCY EQUIPMENT 



(5.1) General Layout 

The operational experience obtained 
on the two experimental recording equip- 
ments originally installed at Alexandra 
Palace led to the conclusion that a re- 
cording suite, intended for the continuous 
recording of programmes, should consist 
of three channels. This would allow one 
channel to be in use while the second 
channel was being loaded ready for 
change-over, the third channel acting 
as a spare. In the event of breakdown, it 
should be possible to have the spare 
channel in operation within two or three 
seconds. The three recording monitors 
should be adjusted and ready for use at 
all times, and, should a change be made 
in the operating conditions during a 
recording, it should occur simultaneously 
on all three monitors. 

At the Lime Grove television studios, 
the signals to be recorded are passed into 
a vision channel where the synchronizing 
signals are first removed. Control of 
gain, lift, frequency correction and con- 
trast-gradient correction are then ap- 
plied. The corrected vision signals and 
separate line- and frame-synchronizing 
signals are passed to distribution ampli- 
fiers which feed the recording picture 
monitors or display units. The electrical 
controls for the vision channel are cen- 
tralized at a control desk, which has 
built-in picture- and waveform-monitor- 
ing facilities. It is possible to observe 
the incoming picture on a white cathode- 
ray tube and to compare this with a 
duplicate of the picture appearing on the 
recording cathode-ray tubes, provided 



by the additional display unit No. 4. A 
spare vision channel is provided. 

Before the start of a recording, test 
waveforms are applied to the vision 
channel in use and the local amplitude 
and cathode-ray-tube bias controls on 
each display unit are adjusted to give the 
same transfer characteristic. Measure- 
ments are made using a multiplier photo- 
cell mounted in a probe held in contact 
with the glass surface of the cathode-ray 
tube, so that the brightness of a small area 
of the screen can be taken. A typical 
test waveform produces a picture of 15 
squares, each of a different brightness. 
Five squares are disposed horizontally 
and three vertically. The amplitude of 
the voltage corresponding to each square 
varies between zero and 100% of peak- 
white modulation, in linear steps. By 
measuring the brightness of each square 
in turn, the picture-monitor transfer 
characteristic can rapidly be checked and 
made similar on the four display units. 

To facilitate change-overs the camera- 
motor controls and the sound and vision 
faders are situated on the control desk. 

(5.2) Electrical Correction 

The electrical corrections introduced 
in the vision channel are as follows : 

Frequency Correction: To compensate for 
the loss of fine detail introduced by the 
camera optics and film, a high-frequency 
lift is applied in the vision channel. Since 
the losses can be expressed as an aperture 
effect, the amplitude of the lift is made 
approximately equal to the square of the 
frequency. A typical value of lift would 
be 9 db at 3 me. 



W. D. Kemp: Television Recording Abstract 



379 




VACUUM 



Fig. 8. An enlarged view of the 
cathode-ray-tube screen. 



Phase Correction: This is required because 
the frequency correction tends to introduce 
some additional phase distortion, and also 
to make more apparent phase errors 
occurring earlier hi the chain of equipment 
preceding the television-recording system. 

Contrast Gradient Correction: Owing to the 
density /log-exposure characteristic of the 
film it is necessary to overemphasize the 
tone separation in the highlights and 
darker tones of the picture to obtain 
correct tone reproduction on the film. 
By the use of a nonlinear amplifier, the 
contrast gradient is increased by a factor 
of about 2:1 (see Sec. 6). 

Lift Correction: For a television picture of 
full tone range, the difference between 
the voltage corresponding to the darkest 



tone in the picture (picture black) and 
the voltage corresponding to the tip of the 
synchronizing pulses ( black level) is known 
as the lift. For television signals derived j 
from most cameras, this is varied by a 
manual control manipulated by the camera 
operator. The contrast-gradient correc- 
tion employed renders the vision channel j 
particularly susceptible to changes in lift, 
and a control is therefore provided to .1 
enable minor adjustments to be made. 

(5.3) The Display Units 

To minimize the loss of quality in a 
television-recording system, it is essential < 
that the recording monitor should be of 
the highest attainable standard. Fine 
focus, maintained over the whole screen 
even at peak modulation, is particularly 
important. A substantially perfect inter- 
lace is also essential. Freedom from drift 
in cathode-ray-tube bias and drive over 
long periods of time is a necessity, since 
no adjustments to individual display 
units can be made once a recording has 
commenced. The geometrical distor- 
tion introduced by the recording monitor 
must also be extremely small. 

The vision frequency equipment in- 
stalled at the Lime Grove studios was 
built to a B.B.C. specification covering 
the above points. 15 The response be- 
tween 10 kc and 3 me was specified as 
constant within 0.5 db. 

(5.4) Cathode-Ray Tubes for Recording 
Purposes 

Film emulsions, even those of the 
panchromatic type, tend to possess maxi- 
mum sensitivity at the blue end of the 
spectrum. This alone would suggest the 
choice of a blue phosphor such as silver- 
activated zinc sulfide for the recording 
cathode-ray tube, and this choice has 
the following additional advantages: 

(a) The phosphor can be made with a 
very fine grain structure. 

(b) The efficiency is high. 

(c) The afterglow is short. 

(d) The use of a single colour helps 
optical resolution as the lenses used need 
not be so accurately colour-corrected. 



380 



April 1953 Journal of the SMPTE Vol.60 



(e) Screen saturation occurs only at a 
high brightness level. 

Silver-activated zinc sulfide has there- 
fore been used both here and in America 
for television recording. With recording 
processes involving relative movement of 
the film and image, the decay time of the 
phosphor is more important than colour 
or efficiency, and zinc-oxide mixtures 
might be of advantage. 

The most suitable size of cathode-ray 
tube is governed by a number of factors : 

(a) The ratio of screen size to scanning- 
spot size should be a maximum. This 
discriminates against the smaller cathode- 
ray tubes, such as those of 5-in. diameter 
or under. 

(b) The brightness should be high. 
For a given beam current and accelerating 
potential, the screen brightness is inversely 
proportional to the area of the raster, 
which favours the use of a small cathode- 
ray tube. 

(c) Blemishes and grain structure in the 
phosphor should be negligible. Since 
these tend to be of constant size, inde- 
pendent of screen diameter, it is desirable 
to have as large a screen as possible. 

(d) The curvature of the cathode- 
ray-tube face should be slight. 

Bearing the above factors in mind a 
9-in. cathode-ray tube with an optically 
flat screen would seem to be a good 
compromise. This size was ultimately 
chosen for the display units used in the 
Lime Grove recording suite. The raster 
size was 6.7 in. X 5.02 in. 

The effect of light reflections in the 
cathode-ray tube is worth consideration. 
Reflections may be divided into three 
types, as follows : 



(a) Reflections from glass surfaces other 
than those of the screen. 

(b) Reflections which cause a haze over 
the whole surface of the screen. 

(c) Reflections which cause a halo round 
a highlight in the picture. 

Reflections of type (a) can be elimi- 
nated by the use of aluminium backing 
for the phosphor. This is desirable in 
any case, since the light output is in- 
creased and the danger of ion burn re- 
duced. The crystals of the phosphor 
emit light in all directions, and the 
aluminium coating tends to reflect back 
that portion of the light emitted in the 
direction away from the recording 
camera while considerably attenuating 
any light which may have penetrated the 
coating and been reflected from glass 
surfaces behind the screen. 

Reflections of types (b) and (c) are 
less easy to reduce as they are due to the 
vacuum/glass and air/glass surfaces of 
the screen itself. Figure 8 (Fig. 9 in the 
complete paper) shows an enlarged view 
of the cathode-ray-tube screen. From 
any point P, rays of light PR, PA and 
PC are emitted. The portion of the ray 
entering the glass will again be par- 
tially reflected at C and again at D, 
E and so on. Such rays near the critical 
angle give reflections of type (b). 

Reflections of type (c) arise from rays 
such as PA, which is partially reflected 
at A back on to the phosphor at B. The 
halo is intensified by light travelling from 
P in the direction of the arrows through 
the phosphor itself. 

A possible way of reducing type (b) 
and (c) reflections would be to make the 
glass attenuate the light. 



(6) TONE REPRODUCTION 



The television camera, and its associ- 
ated apparatus, converts the light and 
shade of the original picture into elec- 
trical signals of particular amplitude. If 
these signals are used to modulate a 
cathode-ray tube, and to expose photo- 



graphic film, then, when the developed 
and printed film is transmitted on a 
telecine machine, the electrical signals 
produced should exactly correspond in 
amplitude to the original signals. This 
would ensure that the tone reproduction 



W. D. Kemp: Television Recording Abstract 



381 



of the television recording would be 
identical to that of the original and repre- 
sents an idealized case. 

Unfortunately the tone reproduction 
of the original television camera varies 
according to the camera tube employed 
and in most cases is far from perfect. 
The B.B.G. has in use six types of 
camera. If correction is applied it 
should be done in two steps, the first one 
in the camera chain to obtain better tonal 
characteristics for direct transmission, 
and the second step in the recording proc- 
ess to compensate for its deficiencies. 

At the present time, owing to signal/ 
noise ratio, absence of black reference 
level and other considerations, it has not 
been found possible to correct each type 
of camera to give a standard transfer 
characteristic. 

(6.1) Transfer Characteristics of Record- 
ing Process 

Six transfer characteristics are involved 
as follows: 

(a) Amplitude linearity of the electrical 
equipment associated with the display unit. 

(b) Transfer characteristic of the 
cathode-ray tube used for recording. 

(c) Optics of the recording camera. 

(d) Overall characteristic of the nega- 
tive stock and its development. 

(e) Grading of the print. 

(f) Type of printing stock and its de- 
velopment. 

(a) The amplitude linearity of elec- 
trical equipment can be made very good 
indeed. The B.B.C. specification for the 
Lime Grove 35mm installation specified 
a limit of 2% up to 100% modulation 
and an additional limit of 5% up to 
150% modulation. The contrast-cor- 
rection amplifier provided, however, con- 
siderably modifies this response (see Sec. 
6.3). 

(b) The transfer characteristic of a 
recording picture monitor may be taken 
by applying a special test waveform as 
described in Sec. 5.1, or by applying a 
number of waveforms in turn, each giv- 



ing a uniform screen brightness of dif- 
ferent value. 

(c) The lens system of a recording 
camera may give rise to multiple reflec- 
tions from the lens surfaces themselves 
and from the lens barrel. These produce 
a naze over the picture which affects the 
slope of the transfer characteristic, par- 
ticularly in the dark regions. 

(d) The normal density/log-brightness 
characteristic of the negative stock is 
somewhat modified by the short exposure 
period obtained in television recording. 
Even with a long-afterglow phosphor 
this is short by photographic standards, 
and the law of reciprocity breaks down. 
The effect is to sharpen the toe of the 
curve, which is an advantage since the 
straight portion is slightly increased. 

The development control gamma of 
the negative must be such that the over- 
all mean contrast gradient of the record- 
ing process is approximately unity. 

(e) The grading of the print follows 
normal motion picture practice, less fre- 
quent shot-to-shot grading being re- 
quired, since this is controlled by the tele- 
vision camera operators. 

(f) A standard motion picture tech- 
nique is used in order to simplify the 
speed and ease of obtaining copies. The 
electrical contrast correction in conjunc- 
tion with the negative development pro- 
vides a negative of standard characteris- 
tics. 

(6.2) High-Gamma Recording 

The contrast range which can be 
accommodated in the recording cathode- 
ray tube is limited by the various reflec- 
tions which occur (see Sec. 5.4). If the 
contrast range is reduced by lowering the 
contrast gradient electrically, then the 
effect of these reflections is minimized. 

In the Lime Grove recording installa- 
tion the contrast-correction amplifier 
reduces the electrical contrast gradient to 
approximately 0.4 over the mid-tone 
range, and allows an increase of this 
figure to about 0.8 in the highlights and 
shadows. The negative control gamma 



382 



April 1953 Journal of the SMPTE Vol. 60 



can then be raised from about 0.7 to over 
unity, with consequent increase in nega- 
tive sensitivity. 

Additional sensitivity is also obtained 
from the change of contrast gradient, 
since if the highlights are kept at the 
same brightness, any particular tone in 
the picture will be brighter. 

(6.3) Direct Positive Recording 

If the vision signals are reversed in 
polarity before being applied to the re- 
cording cathode-ray tube, a negative 
picture is produced. Unfortunately, 
however, this picture no longer bears a 



simple-power-law brightness relation to 
the original scene. 

If a reciprocal law is introduced into 
the vision chain, the simple-power-law 
relationship can be re-established and a 
single photographic process will yield a 
positive. Provision for this type of re- 
cording is made in the Lime Grove instal- 
lation. 

Such "direct positives" have been 
widely used in the United States, mainly 
for economic reasons. In most cases, a 
simple polarity reversal is employed, and 
the tonal distortion tolerated. 



(7) CONCLUSIONS 



Since the overall resolution of television 
on present standards is low compared 
with optical systems and 35mm motion 
picture film, it might be thought that the 
I resolution of these items would not be a 
limiting factor in the overall definition 
of a television recording. Experience 
has shown that this is not the case and 
that it is essential to have first-class opti- 
cal systems and fine-grain film stock for 
best results. When 16mm film is used 
this is even more essential. 

Interlaced recording by continuous 
motion implies a very high degree of 
velocity stability and usually necessitates 
complex optical systems. For these rea- 
sons it would appear that intermittent- 
motion methods are more likely to pro- 
vide the ultimate solution. The develop- 
ment of a suitable quick-pull-down 
method is therefore of primary impor- 
tance, since a simple optical system can be 
used and difficulties due to cathode-ray- 
tube afterglow do not arise. 

There is room for improvement in the 
cathode-ray tube used for recording. 
It is felt that existing designs are based 
largely on normal television-receiver 
practice, where cost is of primary impor- 
tance. It may be that by reducing the 
life of the cathode-ray tube, or some 
other such factor, considerable improve- 
ment could be obtained. 



Much work remains to be done on con- 
trast-gradient correction and general 
photographic technique. Further gains 
can be expected in this direction in the 
future. 

The recording of television programmes 
is now a standard operational procedure 
both here and in America. Since most 
of the development has been done since 
the war, this must be considered the 
youngest branch of television. There 
seems to be no reason why further de- 
velopment should not yield results of a 
standard comparable with that of sound 
recording. If this is so it may be ex- 
pected that the scale of television record- 
ing will be increased to provide a com- 
parable number of programme hours. 

For countries starting television sys- 
tems, it seems that cost will necessitate a 
large percentage of recorded programmes, 
and the further development of 
television-recording technique for the 
interchange of programmes therefore 
seems of international importance. 

Acknowledgments 

The author is indebted to the Chief 
Engineer of the British Broadcasting 
Corporation for permission to publish the 
paper, and to W. H. Cheevers and B. R. 
Greenhead for help in its preparation. 
He also wishes to thank others who have 



W. D. Kemp: Television Recording Abstract 



383 



contributed information on which the 
paper is based, notably H. Bastie and 
L. Cheeseman (E. F. Moy, Ltd.), W. 
Vinten (W. Vinten, Ltd.), T. C. Nuttall 
and G. S. Freeman (Cinema Television, 
Ltd.), T. C. Macnamara (High-Defini- 
tion Films, Ltd.), A. B. Howe (B.B.C. 
Research Dept.) and Dr. H. H. Hopkins. 

References 

1. A. D. Blumlein, G. O. Brown, N. E. 
Davis and E. Green, "The Marconi- 
E.M.I. Television System," J. Inst. 
Elec. Engrs. (London), 83: 758, 1938. 

2. P. H. Dorte, Unpublished B.B.C. 
correspondence, 1948. 

3. H. W. Baker and W. D. Kemp, "The 
recording of television programmes," 
B.B.C. Quarterly, 6: 1, Winter 1949-50. 

4. G. Vinten, "Image super-imposition," 
Brit. Prov. Pat. 284548. 

5. B.B.C. and J. A. Fitzgerald, "Improve- 
ments in and relating to the recording 
of television signals," Brit. Prov. Pat. 
Appl. 18727, 1950. 

6. R. M. Fraser, "Motion picture photog- 
raphy of television images," RCA 
Rev., 9: 202, 1948. 



7. D. A. Smith, "Television recording," 
Wireless World, 55: 305, 1949. 

8. B.B.C., C. Duddington and W. D. 
Kemp, "Improvements in and relating 
to the recording of television pictures 
upon cinematograph film," Brit. Pat. 
Appl. 23216, Sept. 1950. 

9J 1 W. D. Kemp, "A new television re- 
cording camera," J. Brit. Kinemat. 
Soc., 27: 39-46, Aug. 1952. 

10. F. N. Gillette, "The picture splice as 
a problem of video recording," Jour. 
SMPE, 53: 242-255, Sept. 1949. 

11. F. N. Gillette and R. A. White, "New 
video recording camera," Jour. 
SMPTE, 56: 672-679, June 1951. 

12. C. R. Fordyce, "Improved safety 
motion picture film support," Jour. 
SMPE, 51: 331-350, Oct. 1948. 

13. J. M. Calhoun, "Physical properties 
and dimensional behavior of motion 
picture film," Jour. SMPE, 43: 227- 
266, Oct. 1944. 

14. T. C. Nuttall, "More about spot 
wobbles," Wireless World, 56: 189, 1950. 

15. J. E. B. Jacob, "High performance 
television monitors," J. Brit. Inst. 
Radio Engrs., 10: 158, 1950. 



384 



April 1953 Journal of the SMPTE Vol. 60 



Synchro-Lite Powered 16mm Projector 

By R. E. PUTMAN and E. H. LEDERER 



A new flashtube arrangement, providing projection light for television repro- 
duction from film, is here described in detail. It assures accurate and perma- 
nent synchronization of light pulse with intermittent pulldown. Further, 
since both the intermittent movement and the flashing lamp are controlled by 
the television vertical pulse, the mechanism can readily be interlocked with 
other picture sources for interpolation, laps and fades. Travel ghost resulting 
from motion of film while the light pulse is on is made impossible by the design. 



IVJLoTiON PICTURE film plays a very 
important part in television stations, 
both as to programming and as to reve- 
nue. The projection equipment used, 
therefore, deserves the most serious 
consideration and attention. Methods 
and devices developed for the very dif- 
ferent purpose of projecting a motion 
picture before an audience are not well 
adapted to energizing an iconoscope. 

The value of pulsed light for film pro- 
jection in a television system has long 
been recognized. Among its advantages 
are very long lamp life, still frame pro- 
jection without risk of damage to film, 
accurate control of width of light pulse, 
high iconoscope output, and freedom 
from phasing bar in the picture area. 

The new Synchro-Lite* Projector to 



Presented on May 9, 1952, at the Atlantic 
Coast Section Regional Meeting at Atlanta, 
Ga., by R. E. Putman, who read the paper, 
and E. H. Lederer, Broadcast Studio Engi- 
neering Section, Electronics Div., General 
Electric Co., Electronics Park, Syracuse, 
N. Y. This article is part of a paper pre- 
sented at the Sixth Annual NARTB Con- 
ference, Chicago, April 1952. 
* Synchro-Lite is a General Electric trade- 
mark. 



be described here was designed to offer 
additional advantages. Performance, ac- 
cessibility and simplicity have been im- 
proved. The design permits accurate 
and permanent synchronization be- 
tween light pulse and projector action. 
Drift from synchronization is eliminated. 
Possibility of travel ghost, resulting from 
film transit during the interval the light 
pulse is on, has also been eliminated. 
Interlock between this projector and 
other picture sources, including sources 
remotely located, can be accurately 
maintained so far as this projector is 
concerned, since its entire action is con- 
trolled by the television vertical pulse. 

Mechanism and Lamphouse 

The projector mechanism is that de- 
scribed by Frittsf with some relatively 
minor modifications. One is elimina- 
tion of the projector shutter, which is 
not needed, of course. Another is a 
change in the constants of the coiled 
spring in the tuned coupling between the 

f Edwin C. Fritts, "A heavy-duty 16mm 
sound projector," Jour. SMPTE, 55: 425- 
438, Oct. 1950. 



April 1953 Journal of the SMPTE Vol.60 



385 



PULSE 
AMP 




Fig. 1. Circuitry of the Synchro-Lite. 



1440-rpm motor and the intermittent 
movement. This Fritts described as 
tuned to produce a normal 57 pulldown 
adjustable to somewhat less than 50 
for television. We have adjusted it to 
52. This produces a pulldown time 
of 6000 /zsec, which in turn allows a 9 
phasing tolerance based on the maximum 
vertical blanking time of 8%. 

A more drastic departure from the 
projector as described by Fritts is of course 
in the lamphouse, where an entirely 
different light source is used. The hous- 
ing is cast aluminum with outside adjust- 
ments for the height of the flashtube and 
position of the reflector. The dual 
aspheric lens condensing system is of 
the slide type, readily removable for 
cleaning. This condensing system was 
specially designed to provide an even 
and high degree of illumination on the 
film aperture from the small light source 
of the tube. The standby projection 
lamp and provisions for its immediate 
substitution in case the lamp in opera- 
tion fails, one of the features of the 
Fritts design, will not be found in this 
unit since the flashtube used, GE FT-231, 
fails gradually after hundreds of hours 



of use and may be replaced at leisure. 
Two 5 J-in. diameter fans provide forced 
air cooling for this unit. The power 
chassis is vibration-insulated from the 
pedestal to protect the sound system 
against any iron core vibrations. 

Circuit Analysis 

Electrical arrangements of the circuit 
are indicated in block schematic in Fig. 1 . 
The circuitry there shown performs five 
functions: (1) triggers the flashtube in 
synchronism with the vertical driving 
pulse; (2) provides power for the flash- 
tube; (3) controls the width of the 
resultant light pulse by extinguishing 
the flashtube after the predetermined 
interval of 830 ^sec; (4) protects the 
flashtube against damage that might 
otherwise result from severe noise on the 
pulse line or other synchronizing genera- 
tor trouble; and (5) synchronizes the 
projector motors with the flashtube 
action. 

There are two thyratrons in the circuit: 
a 2D21 which is part of the trigger; 
and a 5545 which extinguishes the pulse. 

Trigger action begins at the upper 
left of Fig. 1, where the vertical driving 



386 



April 1953 Journal of the SMPTE Vol. 60 



pulse is amplified and applied to the 2D21 
thyratron. This tube can be regarded 
as a one-kick blocking oscillator. Its 
plate circuit contains an autotransformer 
that produces an output pulse 3 /xsec 
wide and 4000 v in amplitude. This 
pulse of high voltage is used to ionize 
Jthe gas in the flashtube. 

The power supply unit, driven by a 
standard 117-v a-c line, is composed of 
selenium rectifiers in a full-wave circuit, 
delivering 150 v d-c at approximately 
2 amp. This d-c output is inadequate 
for the requirements of the flashtube if 
used as d-c, but since it is wanted for 
only 830 //sec it can be readily peaked 
into an adequate supply. This is ac- 
complished by LI and Cl of Fig. 1. 
These two components constitute a 
resonant circuit of a frequency of 30 
cycles/sec. It discharges its energy 
through the flashtube at a peak current 
jof 70 amp and a peak voltage of 500 a-c, 
Iplus the power supply's 150 v d-c. 

The thyratron, 5545, is fired simul- 
taneously with the flashtube; and the 
discharge of Cl completes its circuit 
through the 5545. LI, however, ap- 
plies a negative bias to the plate of the 
5545 after 830 jusec. The 5545 is thus 
cut off, and the cycle of events associated 
with one flash of the flashtube is thereby 
completed. 

The protective circuit of Fig. 1 , consist- 
ing essentially in the 12AT7 tube, 
operates through a plate circuit relay to 
open the line to the selenium rectifier. 
The tube is normally biassed to cut off; 
however, if the grid voltage of the 5545 
thyratron drifts positive from its nominal 
value of 17, the 12AT7 conducts and 
its plate circuit relay is energized. 

The filter capacitors of the selenium 
rectifier circuit provide the 60-cycle a-c 
used for driving the intermittent motor. 
The power available is ample for that 
purpose; and, since the phase of this 
voltage is dependent on the firing of the 
flashtube, motor and light are always 
locked together. In effect, the motor is 
controlled by a remotely located syn- 



chronizing generator. (When a still is 
projected, an inductance-resistance net- 
work is substituted for the motor, pro- 
viding the same load on the filter as if 
the motor were running and thus keep- 
ing the flashtube current unchanged.) 
The second harmonic distortion in the 
filter-capacitor waveform is not of sig- 
nificant amplitude in this operation. 
Capacity coupling is used between motor 
and filter. 

Results Obtained 

The spectrum and the low duty cycle 
of the flashtube make its output a cold 
light. A spectral analysis of the output 
of the FT231 flashtube shows that it 
has its main peak at about 4600 A. A 
second peak, which loses its pulsed char- 
acter, rises in the infrared region, 
mainly because the electrodes operate 
at an incandescent temperature. To 
prevent this infrared light from reaching 
the iconoscope, a blue-green filter is used. 

Curiously, despite the fact that the 
filter introduces a loss of approximately 
20% of the total light energy-, there is 
normally an increase in the signal output 
from the iconoscope when the filter is 
introduced. The same result follows 
if the filter is used with an incandescent 
lamp. A number of theories have 
been advanced to explain this phenom- 
enon, but at the moment we do not 
know of any explanation that is entirely 
satisfactory. Corning 9780 and 9788 
are the two filters in current use. 

Comparison of the light output from 
this projector using the FT231 flashtube 
with that delivered by a 1000-w incan- 
descent operating through a shutter is 
difficult to draw because of lack of satis- 
factory standards. The Joint SMPTE- 
RTMA Film Equipment Committee, 
TR4.8, is working toward a solution. 
They are at present considering, but 
have not as yet determined upon, a 
filtered light meter calibrated in Icono- 
scope Exposure Units. That is, just as 
foot-candles are measured by a detector 
having a spectral sensitivity similar to 



Putman and Lederer: Synchro-Lite Projector 



387 




388 



Fig. 2. The complete projector. 
April 1953 Journal of the SMPTE Vol. 60 



[that of the human eye, lEU'S would be 
measured by a detector having a spectral 
sensitivity similar to that of an 1850-A 
iconoscope. At 2700 K the foot-candle 
meter and the IEU meter would give 
(identical readings. 

Another way to evaluate the light 
butput of a particular projector is to 
use a standard television iconoscope 
camera chain adjusted for optimum 
pictures, and compare the maximum 
signal output from this projector with 
that from a similar mechanism using an 
ncandescent light source . Several checks 
have been made by this method. They 
indicate that there is essentially no dif- 
jference between an FT231 flashtube 
(operating as herein described and a 
1000-w incandescent lamp operating at 
Jts normal 1 1 5 v. 

Figure 2 shows the complete projector. 



Specifications for the General Electric 
Projector PF-5-A 

PICTURE 

Lens: 4-in. focal length, //1. 5. 

Image distance for 3f X 4\ image: 55 in. 

from aperture. 
Resolution of lens: 90 lines/mm over a flat 

field. 
Dynamic resolution: 60 lines/mm or 850 

television lines. 
Light source: FT231 Flashtube. 
Steadiness: 0.15% jump and 0.08% weave. 

SOUND 

Flutter: less than 0.2% rms. 
Equivalent slit width: 0.0003 in. 
Signal-to-noise ratio: 55 db, using SMPTE 

400-cycle Signal-Level Test Film. 
Output: + 14 dbm. 
Output impedance: 150/600 ohms. 
Distortion: 0.5%, 50 to 6000 cycles/sec. 

GENERAL 

Tilt: 10% from horizontal. 
Power input: 700 w. 
Weight: 450 Ib. 

Size: 26 in. long, 20 in. wide, 57 in. high 
(less upper reel). 



Putman and Lederer: Synchro-Lite Projector 



339 



New Professional Television Projector 



By W. E. STEWART 



A new professional projector specifically designed to meet television needs 
which features a high-fidelity sound system with fast stabilization time, i< 
described. A 2-3 pulldown system is incorporated especially for television 
All gearing runs in oil and gears and other mechanical parts are designed foi 
long life. Projection lamps change automatically in the event of filament 
failure. Still pictures can be shown. 



HEN TELEVISION first came on the 
air, the exact role that film would play 
in program presentation was not well 
known. The convenient and simple 
method for obtaining film programs 
was to adapt existing projector designs 
to the job. These projectors have gone 
through an evolution as they were 
adapted more and more closely to the 
special demands of television. 

During this same period, we have had 
the opportunity to see the trend so far 
as program material is concerned and to 
confirm the belief that film is an ex- 
tremely important part of the program. 
Much of the income for both large and 
small stations will be derived from pro- 
grams that go on the air from film. 

Operating techniques have also gone 
through some evolution and are shaking 
down into a reasonably stable pattern. 
It appeared appropriate therefore to 
design a new projector specifically for 
television, and the new RCA professional 



Presented on October 6, 1952, at the Soci- 
ety's Convention at Washington, D.C., by 
W. E. Stewart, Radio Corporation of 
America, RCA Victor Div., Engineering 
Products Dept., Camden 2, N.J. 



television projector, Type TP-6A, for 
16mm film is such a machine. 

Before starting to design, the oppor- 
tunity was available to look over general 
requirements and decide whether the 
equipment might take some entirely 
new form. It appeared that the follow- 
ing general objectives should be included: 

1 . The projector should be a building- 
block type of item. This means that it 
should be one able to fit into existing 
systems and into as many combinations 
as possible for future systems. It should 
be separate from optical multiplexing and 
camera equipment, 

2. The main working parts should be 
at a convenient height for a man standing 
at the machine, to provide facility in the 
putting in and taking out of film, and 
for operation and servicing. In other 
words, a pedestal is required. 

3. The equipment should be adapted 
to television programming. It should 
therefore include mechanism for quick 
and easy threading, and for fast starting 
and stopping. There should be a mini- 
mum of time loss from burned-out lamps 
and other causes. The projector must 
be able to show still pictures and be capa- 
ble of operation by remote control. 



390 



April 1953 Journal of the SMPTE Vol. 60 




Fig. 1. RCA Type TP-6A television projector. 
W. E. Stewart: Television Projector 



391 




Fig. 2. Automatic projector 
lamp change mechanism. 

The factors mentioned above are of a 
comparative nature, and it takes con- 
siderable judgment to determine where 
to put the emphasis. As can be seen 
from Fig. 1 the machine has a general 
resemblance to other projectors since it 
stands on a pedestal and can be serviced 
from all sides. The extent to which the 
design engineers stressed various fea- 
tures will become more obvious as details 
are explained. 

Figure 2 shows the automatic projec- 
tor lamp change mechanism. This 
mechanism consists of a turret arrange- 
ment which holds a 1000-w projector 
lamp accurately in position until there 
is a filament failure. The moment the 
filament current is interrupted, a motor 
swings the turret around 180 and a 
new projection lamp is swung into place. 
This operation requires approximately 
one second and assures a minimum of 
interruption to a television program from 
this common cause of failure. A pilot 



Fig. 3. Removal of lens assembly. 



lamp shows when the bulb has been 
changed, and the spare can be replaced 
while the machine is in operation. 

A new high-efficiency condenser lens 
system has been designed for this projec- 
tor. It includes a filter for taking out the 
red and infrared light so that a sharper 
television image results. Removing red 
also gives a better monochrome picture 
with color film. The lens assembly 
is easily removed, as can be seen in 
Fig. 3. 

Two motors are used in the drive 
system. One handles the shutter while 
the other drives the sprockets and inter- 
mittent. Both are 115-v, single-phase, 
synchronous motors. These allow still 
pictures to be shown, an important con- 
sideration in television studio techniques. 

In order to get as much light as possi- 
ble on the film camera, a new, fast-pro- 
jection lens with speed of //1. 5 has been 
designed. This lens, made for the short 
throw normally found in television ap- 



392 



April 1953 Journal of the SMPTE Vol. 60 




Fig. 4. Interior of projector, showing lens position. 



plications, has a focal length of 3j in. 
Fhe lens is held in a yoke with a special 
inti-backlash mounting. Focus con- 
rols are available on both the front and 
3ack of the projector, so that it is possible 
o focus the system from either side while 
ooking into the film camera. The pro- 
ector lens is so mounted that it is not 
iccessary to move it while threading the 
ilm, which also insures that it remains in 



focus. There is a built-in iris in the pro- 
jector lens. Figure 4 shows this lens 
just under the pointing finger. 

The same figure gives an excellent 
view of most of the film path. Large 
or small reels may be used on this pro- 
jector. A 4000-ft reel which holds 
enough film for one hour and 50 minutes 
is shown, but small reels with short 
commercials on them may be used just 



W. E. Stewart: Television Projector 



393 




o 



Fig. 5. Path of film 
through sound sys- 
tem. 










394 



Fig. 6. Interior of projector, showing sound system. 
April 1953 Journal of the SMPTE VoL 60 



as easily. The supply- reel shaft at the 
top has a friction brake with a broad 
noncritical adjustment. 

The machine features easy threading. 
In the figure the finger is shown on a 
lever which swings the pressure shoe 
away from the gate so that the film can 
be slid quickly into place. 

There is an adjustment on the pressure 
shoe so that pressure may be changed 
while the projector is running, according 
to the type of film in use. The knob and 
small calibration dial may be seen di- 
rectly under the thumb in Fig. 4. 

A new sprocket shoe has been devel- 
oped which can be swung away from the 
sprocket for cleaning but which will 
always assume proper alignment and 
not jam against the sprocket teeth. 
For threading, the front edge is simply 
pulled outward while the film is slipped 
into place. 

The sprocket and intermittent mech- 
anism may be turned by a knob at the 
top to check the threading and to ad- 
vance the film one frame at a time. 

A lamp at the bottom supplies light 
for threading in darkened projection 
rooms. The clear plastic door keeps 
dust from the moving parts but allows 
constant surveillance of the film. 

A great deal of study went into the 
pulldown mechanism. There is some- 
times prejudice against the claw pulldown 
used here because it has been associated 
with the simpler and cheaper forms of 
projectors. Careful analysis of the mech- 
anisms, however, indicates that there is 
much more opportunity for accurate 
film indexing in such a simple device, 
which goes through exactly the same 
cycle for each frame of the film, than 
there is in the sprocket pulldown system 
with its multiple indexing positions for 
the components of the system in suc- 
cessive frames. A high degree of 
picture stability has been attained, and 
should be easy to maintain, in the manu- 
facture and use of this mechanism. It 
is also important that a lost loop can be 
restored when using the claw pulldown. 



A three-tooth claw is used in this mech- 
anism, so that damaged film will go 
through it with a minimum of difficulty. 
The upper tooth and the movable side 
guides are sapphire-lined. All the gears 
and other drive mechanism, except the 
parts which must be outside to move the 
film, run in oil and are housed in a 
sealed case. All life tests made indicate 
that thousands of hours of service can be 
expected. Complete figures on this 
are not available, as not one of the mech- 
anisms has yet worn out. 

Theatrical-type framing is used for 
the picture. That is, the position of the 
film, rather than the aperture, is moved. 
The framing knob is on top, where it can 
be reached on either side of the projec- 
tor a feature necessary in television. 

Figure 5 shows the path of the film 
through the sound system. The cover 
has been removed so that both exciter 
lamps are shown in place. The lamps 
are 10-v, 5-amp bulbs operated on d-c. 
They incorporate a new type of inter- 
nal-heat diffusing screen which greatly 
reduces darkening with age. A high- 
efficiency optical system, similar to 
that in the RCA film sound recorder, 
gives enough light in the photocell so 
that the lamp can be operated well 
below full voltage. These factors tend 
to give long life to the exciter lamp. In 
addition, an arrangement is provided 
that allows a new lamp to be shifted into 
place immediately by manually operat- 
ing the lever shown on the extreme right- 
hand edge of Fig. 5. 

The film path for the sound system 
incorporates several new features. The 
pressure roller on the sound drum is 
damped through a viscous drag arrange- 
ment to hold the film tight against the 
drum. After the film passes over the 
sound drum, it goes through a double- 
roller system as shown in the figure and 
this also is damped, so that flutter from 
the sprocket is effectively removed 
from the sound-drum motion. Not 
visible in Fig. 5 is the flywheel on the 
back end of the sound-drum shaft. 



W. E. Stewart: Television Projector 



395 




Fig. 7. Control panel. 



When the projector stops, a brake shoe 
is allowed to come against the flywheel, 
and the film is stopped without being 
jammed ahead by the inertia of the sound 
system. When the machine starts, this 
same brake shoe gives the flywheel a 
boost so that the whole system comes up 
to speed rapidly and is stabilized in ap- 
proximately two seconds. This feature 
is especially useful in television work 
where tight programming is involved. 
The whole sound system is shock- 



mounted so that vibration from the 
projector does not cause microphonics 
(Fig. 6). 

The plug-in amplifier provided with 
the projector is built with broadcast- 
quality components and has a response 
to 10,000 cycles. This response may be 
limited by a switch to 7000 or 5000 cycles. 
There is a volume control on the amplifier 
which allows adjustment of the output 
level up to dbm. This output is ap- 
propriate for connecting to the average 



396 



April 1953 Journal of the SMPTE VoL 60 




Fig. 8. Top and front view, showing access to control panels and internal mechanism. 



consolette circuit and is in line with the 
new RTMA standards which call for 
at least 10-dbm output. The stand- 
ard broadcast audio power supply can 
be used to supply one or several preampli- 
fiers. 

Figure 7 shows the control panel. 
From left to right the controls and indi- 
cators are as follows: 

A. The lamp-failure pilot light; 

B. Reset button for A; 

C. Variac control for projection lamp; 

D. Voltmeter for projection lamp; 

E. Elapsed time indicator; 

F. Main circuit breaker; 

G. Start switch (starts the film motion) ; 



H. Still switch starts the shutter 
motor and places the lamp at full 
brilliance) ; 

I. Pilot light associated with start 
switch; 

J. "Remote-Local" switch (extends con- 
trol to the remote control position) ; 

K. Pilot light for this switch; 

L. Stop button ; 

M. Ready switch (activates the control 
circuits, turns on the blower and pre- 
heats the projection lamp). 

As can be seen from the same view, 
servicing has been made easy on this 
machine. The removal of nine screws 
allows both covers to be removed from 



W. E. Stewart: Television Projector 



397 



the projection head. All parts of the 
projection mechanism are then available 
for servicing while still in normal operat- 
ing condition. This figure gives a good 
view of the blower, lamp-change mech- 
anism, and the intermittent housing. 

Figure 8 is a top and front view which 
shows another kind of access to the pro- 
jector. The control panels and the 
internal mechanism are reached by doors 
with fasteners which can be opened 
quickly for minor servicing or adjustment. 

This view also shows the brake adjust- 
ment for the upper reel and the hand 
turnover knob, framing knob, and change- 
over knob on the top of the projector. 

The main features of the machine 



automatic projection lamp change; 

fast-change exciter lamp ; 

fast-stabilizing sound system; 

4000-ft reels; 

still pictures; 

long life ; 

easy maintenance. 

Special acknowledgment is in order 
for the design work of J. J. Hoehn, H. G. 
Wright and R. N. Lipman who so suc- 
cessfully turned an exacting specification 
into the machine described. Stewart 
Pike guided the styling. 

Discussion 

George Lewin (Signal Corps Photo Center} : 
Can the illumination on this machine be 
reduced readily for image-orthicon opera- 
tion? 

Mr. Stewart: Yes, there is a shutter in the 
lens which allows easy reduction in the 
amount of illumination. Also the Variac 
controls the projector lamp intensity. 

Maxwell A. Kerr (Bureau of Ships) : How 
do you change those projection lamps so 
quickly without breaking a filament? 

Mr. Stewart: Well, if you watch it, you 
don't have the feeling that it's being jerked 
around. It's moved around by a motor in 
quite a smooth motion, but it's about 1 sec 
of time. There is no particular mechanical 
jar to it while it's going. It is detented into 
place so that it is exactly in line and in focus. 



A microswitch breaks the circuit until the 
lamp is approximately located. 

Mr. Kerr: How fast is that pulldown 
time? 

Mr. Stewart: The cam gives a 90 pull- 
down at present. We've used faster ones, 
but this is shorter than our TP35. It's suf- 
ficient for the type of application that it is 
usually used in and will give very long wear 
of the cam. 

Mr. Kerr: That's 90 ... 

Mr. Stewart: The cam with two pulldown 
faces rotates at 720 rpm or 1 revolution 
-P2 sec, which gives 24 pulldowns per sec. 
Each pulldown section of the cam takes up 
45. 



/j45 ^, 1 J_ 
V360 X 12 ~ 9 



_ 
96 

This is the same as 90 of a 1440-rpm cam 
usually .used in projector mechanisms. 

Bernie Elias (Film Processing Specialist, 
Asheville, N.C.} : I didn't quite catch just 
how the pulldown gives more accuracy in 
the film movement than the sprocket-type 
movement. 

Mr. Stewart: We could spend a lot of 
time on that, but I'll put it this way. In 
the sprocket-type pulldown you have 
usually around eight moving parts which 
take different positions for each successive 
position of the film. As a result, extremely 
tight accuracies are required on those parts 
with respect to runout and all other dimen- 
sional tolerances in order to be sure that 
the film always stops in exactly the right 
spot. Here we have a cam which goes 
through exactly the same cycle for each 
pulldown of the film. You have only two 
parts two mating parts which affect the 
accuracy of the position of the film; and 
even if they do vary slightly, it doesn't mat- 
ter, because they are doing the same thing 
on every cycle of the pulldown. We have 
spent a great deal of time on the study of 
that and we have both systems. The TP35 
is the sprocket pulldown. Some of the 
others are the other way. In answer to the 
argument that some of the simpler and 
lower-priced mechanisms use the claw 
pulldown, you can also say that some of the 
most expensive equipment in the film re- 
cording business uses the claw-type of pull- 
down also. But we felt, after a very careful 
study of that subject, that the claw pull- 
down not only gave us a chance to make a 



398 



April 1953 Journal of the SMPTE Vol. 60 



good machine, but it gave a better chance 
to have a machine that would work for a 
long time and maintain that performance. 

Eugene Rector (Fox Theatres) : On the claw 
pulldown type, if you lose a loop, some- 
times isn't it possible to pick it up again 
without shutting down, which you couldn't 
do on your regular intermittent movement . 
I know you can do that on the 16mm. 

Mr. Stewart: Yes, you could restore the 
loop without rethreading. The machine 
doesn't have to stop. You can go right on 
with your projection and restore the loop. 

Mr. Rector: You can't do that on the 
other type movement, on the intermittent 
type? 

Mr. Stewart: No, I believe you cannot. 
That's right. 

Ned Brooke (WSAZ-TV, Huntington, Va.): 
I'd like to ask is provision made to preheat 
the filament on the standby lamp? 

Mr. Stewart: Yes, during the time that it 
swings into place. It makes contact almost 
the instant that the change mechanism 
starts to operate and is on a reduced voltage 
of about 25 to 30 volts and then as it swings 
into place it comes up to full voltage. That 
gives you a preheating time. 



George Lewin (Signal Corps Photo Center) : 
I believe you engage three sprocket holes 
when you pull down. Does that increase 
the problem due to shrunk film? 

Mr. Stewart: No, in the case of shrunk 
film I believe the upper sprocket will do 
most of the work and that's the one that's 
lined with sapphire. No, that doesn't in- 
crease that problem at all. 

Anon: Would you tell us what lubricat- 
ing means is used first on the gear side and 
secondly on the film side? 

Mr. Hoehn (RCA Victor Div.): I can give 
that: Of course, all the gears are housed 
in complete-type compartments and are in 
oil all the time. I am referring to the 
mechanism. 

Anon: And on the film side, how about 
the rollers? 

Mr. Hoehn: All the rollers are nylon, 
except for the sound drum roller which, of 
course, is specially damped. 

Anon: Does the nylon roller not require 
lubrication? 

Mr. Hoehn: Nylon rollers never need any 
lubrication. As far as we know, they last 
longer and have less friction than metal 
rollers. 



W. E. Stewart: Television Projector 



399 



High-Speed Photographic Techniques 
for the Study of the Welding Arc 

By I. L. STERN and JOHN H. FOSTER 



The use of high-speed photography, at rates of 3000 to 4000 frames /sec, for 
the investigation of welding arcs is briefly described, with detailed specifica- 
tions. 



I 



N THE PAST five years, high-speed 
photography has become increasingly 
popular as a method for the investigation 
of the welding arc. Conventional meth- 
ods of studying the welding arc by elec- 
trical metering devices and random 
observations have not been entirely 
satisfactory. In many cases the only 
knowledge of the mechanisms of metal 
transfer and the factors influencing arc 
stability have been derived from infer- 
ences drawn from electrical measure- 
ments. With the advent of high-speed 
photography, the investigator has been 
able to observe the actual physical 
changes hi the welding electrode, arc 
stream and molten weld pool, thereby 
developing information which can be 
considered as fact rather than an infer- 
ence. 



Presented on October 9, 1952, at the 
Society's Convention at Washington, D.C., 
by I. L. Stern and John H. Foster (who 
read the paper), Material Laboratory, 
New York Naval Shipyard, Brooklyn 1, 
N.Y. The opinions expressed herein are 
those of the authors and are not to be 
construed as reflecting the views of the 
Department of the Navy or the Naval 
Service at large. 



For a better understanding of the 
problems involved and the high-speed 
motion pictures (shown at the Conven- 
tion Session), a brief description of the 
welding process will be helpful. Essen- 
tially, the circuit in which arc welding 
occurs consists of a power source, an 
electrode surrounded by a protecting 
atmosphere, the base metal or object be- 
ing welded and the arc stream between 
the base metal and electrode. In some 
forms of arc welding the added weld 
metal is derived from a consumable 
electrode, while in others, a separate 
filler rod, introduced between electrode 
and base metal, is the source of the re- 
quired weld deposit. 

The most common type of arc welding 
is that accomplished with a covered 
metallic electrode. As the arc is struck, 
the metal from the electrode is melted by 
the heat of the arc and transferred 
through the arc stream to the work or 
base metal. As the electrode is con- 
sumed, the covering is progressively 
burned or melted off, the latter forming a 
slag and an atmosphere of gas, which pro- 
tects the molten metal from atmospheric 
contamination, and serves as a source of 
stabilizing ions for the arc. 



400 



April 1953 Journal of the SMPTE Vol. 60 




Fig. 1. Metal transfer during 
short circuit. 



Fig. 2. Explosive burn-off of 
electrode covering. 




Fig. 3. Flaking and uneven burn-off 
of electrode covering. 



The process is accompanied by an 
intense amount of light and heat. The 
fumes from the mineral or cellulosic 
covering envelop the relatively small 
area of action. These plus the intense 
light and extremely rapid and varying 
conditions in the arc complicate the 
problems of the photographer and the 
investigator. 

For practically all investigations of the 
welding arc, only close-up views of the 
largest practicable magnification were 
found to be of value. High-speed pic- 



Fig. 4. Passage of slag from 
melted covering. 



tures other than close-ups do not appear 
much different from films taken at nor- 
mal speeds and do not yield significant 
information. 

In respect to speed required, photo- 
graphic studies have indicated that pic- 
tures taken at 3000 to 4000 frames/sec 
were satisfactory. Frame-by-frame ex- 
amination of the films have in some 
instances indicated that higher rather 
than lower speeds may be required for 
some studies. In several cases, the 
significant changes in the particular fac- 



Stern and Foster: Welding Arc Photography 



401 




Fig. 5. Transfer occurring 
within 1/1000 sec. 



tor under investigation occur in a timd 
interval represented by 1 or 2 frames 
This change would not be evident all 
normal projection speeds and occurs todf 
rapidly for analysis at 4000 frames/seal 
For these special cases, even higher speed) 
Would be desirable. 

In most photographic studies it if 
often necessary to correlate the filnj* 
frames with recorded electrical measure 
ments such as oscillograms. Most highj 
speed cameras are or can be equippec 
with, a flashing argon lamp which form 
light impressions at the film edges a 
intervals of 1/120 sec. The correspond^ 
ing electrical impulse for each flash ma\ 
be used as a reference timing line on tht 
oscillograph. A preferable system is < 
newly developed double-lens camert 
wher v ein the oscillograph trace and the 
visual conditions occurring in the welding 
arc are recorded on a single frame 
simultaneously. 

At the Laboratory with which the 
authors are associated, high-speed pic- 
tures have been taken on black-and- 
white, infrared and indoor- and outdoor- 
type color film. 

Black-and-white film does not differen- 
tiate between the various light fre- 
quencies emitted and consequently the 
resulting pictures are masked by the 
glare surrounding the arc. 

Infrared film tends to overemphasize 
some aspects of the arc at the expense oi 
others and consequently does not give a 
true picture of the arc conditions. How- 
ever, in the case of specialized investiga- 
tions, this distortion of conditions may 
prove desirable because of the over- 
emphasis that it produces. 

Daylight-type color film is the film of 
choice for most investigations of the 
welding arc. Clear, definitive pictures 
of the metallic arc, tungsten arc and 
other welding operations have been 
taken with this film, without the use of 
filters. 

Indoor-type color film may also be 
used; however, in this case appropriate 
filters are necessary for optimum results. 



402 



April 1953 Journal of the SMPTE Vol. 60 



Fig. 6. Reading left, 
then below: Some 
significant changes 
in tungsten elec- 
trode during \ 
cycle of welding 
with alternating 
current (helium 
atmosphere). 




Stern and Foster: Welding Arc Photography 



403 



For metallic arc welding operations with 
covered electrodes a Wratten Filter No. 
29 is recommended. 

The intensity of light emitted by the 
arc is so great that auxiliary lighting is 
considered to be of questionable value. 
Lens openings should be stopped down 
to//22 or lower to prevent overexposure. 
The specifications used for most of the 
films shown at the Convention were as 
follows : 

Film Speed : 3000-4000 frames/sec 
Film: daylight-type, color 
Lens System ://2. 7, 102-mm lens, plus a 

supplementary 220-mm lens of 2-in. 

diam. 

Lens Opening ://22.0 
Distance of forward element lens to arc : 

Tin. 

Auxiliary Lighting: 4 RSP2A bulbs 
Auxiliary Equipment: j hp centrifugal 

blower with 4-in. duct for removal 

of fumes when present 

A completely automatic procedure, 
whereby the camera is actuated at a 
predetermined position of the welding 
electrode, is a practical necessity for 
acquiring the desired angulation and 
view. 

Figures 1 to 5 represent single-frame 
enlargements from the motion picture 



and illustrate the various aspects of the 
welding arc which are shown to greater 
advantage in high-speed motion pic- 
tures. For the purposes of clarification, 
the edges of the welding electrodes have J 
been indicated by dotted lines. Figure 6 
illustrates a tungsten-metal arc process of 
welding with alternating current where- 
by an inert gas such as helium replaces 
the covering as a shielding medium. 
Shown are some of the significant 
changes that occur within f cycle (1/120 ' 
sec). The alternate melting and cooling 
of the tungsten electrode which occur 
during the reversal of current direction 
during the cycle is clearly demonstrated. 

In general, recent developments indi- 
cate that high-speed photographic tech- 
niques will play an increasingly valuable 
role in investigations of the welding arc, 
and in other fields of technology. The 
application of procedures similar to those 
described here should enable the photog- 
rapher to obtain satisfactory high-speed 
motion pictures for welding-arc inves- 
tigations. 

The authors wish to acknowledge the 
cooperation of N. A. Kahn, Head 
Metallurgist of the Material Laboratory, 
for his valuable suggestions and for his 
review of the paper. 



404 



AprU 1953 Journal of the SMPTE Vol. 60 



Use of Photography in the 

Underground Explosion Test Program, 1 95 11 952 



By R. M. BLUNT 



The Underground Explosion Test Program was established to determine the 
resistance of soils and structures to attack with high explosives. This paper 
discusses some of the problems in applying high-speed motion picture photog- 
raphy to the surface phenomena accompanying these tests. Some improve- 
ments in equipment and methods were suggested, based on observations 
made in the field. 



JL HE CORPS OF ENGINEERS, U.S. Army, 
Sacramento District, established a pro- 
gram to determine the effects of under- 
ground explosions in various types of 
soil and rock and on structures within 
these media. This program began at 
Dugway Proving Grounds, Utah, in 
the spring of 1951 and was completed 
last summer (1952) in Buckhorn Wash, 
Utah. Although the primary instru- 
mentation consisted of force gauges of 
various types mounted on the structures 
or in the medium, it was thought 
desirable to supplement the information 
so obtained with high-speed motion 
pictures of the charge and its environ- 
ment. In the spring of 1952, it was 
decided that high-speed motion pictures 
of the interior of the structures would 
provide useful information about the 
behavior of the walls under stress. 
Inasmuch as the characteristics of the 

Presented on October 8, 1952, at the Soci- 
ety's Convention at Washington, D.C., by 
R. M. Blunt, Denver Research Inst., Uni- 
versity of Denver, Denver 10, Colo. 



medium and the location of the charges 
in it were of primary importance to the 
program, the test sites were chosen to 
satisfy the requirements imposed by these 
factors. As a result, the terrain often 
presented considerable difficulty from 
the standpoint of photography. 

The surface of the ground at the 
charge site was usually below the level 
of the surrounding terrain ; consequently, 
in order to photograph the surface of 
the ground above the charge, it was 
necessary either to elevate the camera 
or to bulldoze a long, sloping ramp 
through which the camera could view 
the charge area. However in many 
cases even this was not possible when the 
territory surrounding the charge location 
consisted of a series of small hills, sand 
dunes, tree-covered ground or other 
such natural obstacles. Therefore, in 
many cases the cameras were not 
located at the optimum spot but at the 
best of the few possible locations. The 
rock sites were particularly poor from 
the standpoint of camera locations, 
since there always seemed to be a deep 



April 1953 Journal of the SMPTE Vol. 60 



405 




Fig. 1. Terrain typical of that found in the soil sites. This picture, taken 

from several miles away, under-emphasizes the features which cause trouble 

in finding a good camera location. 



gulley or ravine at precisely the position 
in which one would prefer to put the 
camera (Fig. 1). Much of this difficulty 
might have been overcome by setting 
the cameras very close to the charge 
because a fairly level area usually ex- 
tended for a radius of some 50 ft around 
the charge location. This probably 
would have resulted in a loss of cameras, 
since debris can travel for quite some 
distance at a very high velocity, par- 
ticularly from the charges detonated in 
rock. To reduce the hazard from 
debris, even when set up at some dis- 
tance from the charge, the cameras 
were routinely protected by shelters 
constructed from 2X6 timbers, braced 
by 2 X 4's and 4 X 4's to withstand the 
shock of any flying material that might 
strike them (see Fig. 2). A 3-in. 
circular aperture was cut in the side of 
the shelter which faced the charge to 
permit the camera to see through it. 
The side of the shelter facing away from 
the charge was left open to provide easy 
access to the equipment. The chance 
of anything striking the small target 
represented by the aperture or by the 
lens of the camera was small. We were 



fortunate in this respect, inasmuch as 
no camera was damaged by flying debris 
during the entire program, although 
there were several near misses. 

A difficulty of another type arose 
from the fact that the charges were 
primed in such a manner that the de- 
tonation of the primacord fuse occurred 
directly above the area in which we were 
most interested. The detonation of this 
primacord always left a large cloud of 
black smoke which obscured the early 
portion of the phenomena being photo- 
graphed, particularly when the charge 
was a small one. With a larger charge 
the cloud covered proportionally less of 
the area of interest. 

These difficulties are brought out 
here to provide information for others 
who may set up similar programs. It 
is thought these comments may result 
in consideration of photographic require- 
ments early enough to make proper 
provision for them possible. 

Lighting conditions existing in desert 
country in the West are usually fairly 
good for the purpose of taking high- 
speed photographs between the hours 
of 9 A.M. and 3 P.M. Depending 



406 



April 1953 Journal of the SMPTE Vol.60 




Fig. 2. A rock site installation of camera and power source. 



upon the time of the year, one may 
reasonably expect reflected-light read- 
ings on average countryside to be 
between 800 and 1600 Weston at these 
hours. With this lighting, quite satis- 
factory pictures were obtained at frame 
rates as high as 4000 per second when the 
background consisted of either a light- 
colored sand or a light-colored clay. 
The chief problem that arose was that 
of obtaining sufficient contrast between 
the particular part of the ground in 
which we were interested and the back- 
ground (see the second picture in the 
sequence of Fig. 3). In some despera- 
tion we finally resorted to the expedient 
of spreading powdered coal over the 
surface of the ground immediately 
behind the object we were photograph- 
ing. Since the object seen by the camera 
was of a light color in general this pro- 
vided a reasonable degree of contrast in 
the final photograph. Without such 
precautions it was sometimes extremely 
difficult to distinguish between the rise 
of the ground, in the area of interest, 



and the terrain behind it. When the 
field to be photographed is rather dark, 
as, for example, when the ground is 
covered with sagebrush, it is necessary 
to reduce the frame rate of the camera 
to about 2000 frames/sec when the light 
reading is of the order of 800 Weston 
if satisfactory pictures are to be obtained. 
Due to the requirements of the pro- 
gram, it was necessary to set up the 
high-speed motion picture cameras a 
few hours in advance of the expected 
time of detonation of the charge. This 
was done to allow the photographic 
crew time to set up and man other 
equipment at stations located a mile 
or more distant from the charge site. 
It was thus generally impossible to 
readjust the high-speed motion picture 
cameras for the light conditions existing 
at the time of the detonation. Even in 
the event of considerable changes in the 
prevailing light, as were occasioned by 
storms which frequently arose in the 
late afternoon, we were quite surprised 
at the quality of the pictures that were 



R. M. Blunt: Photography of Underground Explosions 



407 






Fig. 3. A sequence taken from the 



408 




16mm Fastax picture of a typical test. 



409 



taken under these adverse conditions. 
In one instance in particular the light 
reading was approximately 50 Weston, 
the camera was operated at a rate of 
1000 frames/sec, and we still secured a 
picture from which usable information 
could be had. This should not be 
construed as recommending that one 
make a habit of taking pictures under 
such conditions, but merely to indicate 
that it can be done. 

In general, with phenomena of the 
type under investigation in this program, 
it is desirable to set the camera for the 
maximum possible frame rate obtainable 
with the light that is available. This is 
particularly true in rock where the 
phenomena occur so rapidly that a rate 
of less than 2000 frames/sec results in 
pictures which are almost useless from 
the standpoint of analysis. A rate of 
4000 frames/sec is the minimum, and 
it would be better if one could obtain 
pictures at rates more closely approxi- 
mating 8000 frames/sec. However, this 
does not seem possible at the present, 
even with the illumination provided by 
the sun on a bright day in the desert. 
It would be worthwhile to experiment 
with various types of artificial light 
sources to supplement the illumination 
from sunlight. Some means of provid- 
ing a fill light to control contrast would 
make the problem of securing a good 
high-speed picture much easier. In 
this connection it may be well to point 
out that use of Linagraph Pan Film in 
an attempt to secure a higher frame 
speed through the greater emulsion 
speed of this film resulted in quite 
unsatisfactory pictures. We therefore 
carried out a short investigation of the 
suitability of the various types of film 
that we could readily secure (see Fig. 4). 
Unfortunately, the halftone reproduc- 
tion does not convey the differences 
visible in the original films, particularly 
with respect to grain. 

A camera was set up to photograph a 
typical charge area and several different 
types of film were run through in the 






space of some 15 min. During thi 
time the light did not change and t 
camera settings were left the same. 
Therefore, to at least a first approxima- 
tion, the only variable involved was that 
of the film type. This film was all 
processed through the Houston machine 
in D-16 developer at approximately 
80F at about 4 fpm, which would 
correspond to approximately 7^-min 
development time. The film found 
most satisfactory as a result of this 
test was Ansco Gun Film. The rest, jl 
in order of decreasing merit, were: 
Background X, Linagraph Shellburst, ,. 
Linagraph Ortho, Super XX, and 
Linagraph Pan. Of course, one is j 
never satisfied, and it would still be < 
desirable to have film possessing a higher 
emulsion speed than the Ansco Gun j 
Film while retaining the present grain 
and resolution. We are now planning 
an investigation of film plus developer 
combinations that may enable us to 
secure better definition with a reasonable 
film speed. 

As experience was gained in pre- 
dicting the area which could be expected i 
to move following the detonation of the 
charge, it became possible to establish 
a field of view which included only that I 
portion of the ground about the charge 
that would be essential to the analysis. 
This procedure had the obvious ad- 
vantage of giving one the maximum 
amount of information that could be 
obtained from a given picture, but 
since no extraneous elements were 
permitted to appear, the resulting 
pictures are not particularly interesting 
to one who is not endeavoring to make 
an analysis (see Fig. 3). 

For the purpose of making pictures 
that would provide an interesting dis- 
play, it would be better to have the 
cameras set to include the entire explo- 
sion in the field of view. Although our 
major interest lay in the investigation 
of details of the movement of the ground 
around the charge immediately following 
the detonation, some incidental pictures 



410 



April 1953 Journal of the SMPTE Vol. 60 



'were taken to record the motion of 
i Structures near the charge. The thought 
here was that, while the permanent 
displacement of the ground beyond the 
i edge of the crater could be determined 
by other means, the actual motion of the 
structure just following the detonation 
might be much more complicated than 
static measurements could reveal. In 
a similar manner the permanent dis- 
placement of the ground beyond the 
edge of the crater could be determined 
flby measuring the displacement of stakes 
[driven into it. Such stakes were suc- 
cessfully photographed with the result 
Hthat unsuspected details of the motion 
of the ground were revealed. In a 
few cases we also obtained 16mm Koda- 
chrome pictures of the entire phenomena, 
; using the 70-S Bell & Howell cameras 
I Which were set up at a considerable 
|distance from the charge. While from 
jthe analyst's standpoint these pictures 
i provide information only about the 
I growth and spread of the dust cloud 
: originating from the explosion, they are 
[' of considerable value in that they give 
Jan overall picture useful in orienting 
: the observer before looking at the high- 
speed motion pictures made with the 
pastax camera. 

The problem we faced in taking high- 
speed pictures in the interior of the 
.structures was that of supplying sufficient 
i illumination. Suggestions were made 
[that 60-in. searchlights, or flares of the 
type employed by the Air Force, could 
be used. The searchlights, however, 
I require a larger power source than 
j could be moved over the rough terrain, 
I which made them impractical in a field 
i operation of this type. On the other 
Jhand, flares are difficult to control, 
| dangerous to handle, and produce large 
j quantities of smoke. We constructed 
an apparatus consisting of a power 
supply and a motor-driven commutator 
I connected to set off 40 groups of Sylvania 
I2A flashbulbs in sequence (see Fig. 5). 
iTnese bulbs were mounted in reflectors 
) constructed of sheet aluminum along 



the general lines of an overgrown bread 
tin standing approximately 36 in. high 
by 12 in. wide. Each reflector con- 
tained two rows of twenty sockets, a pair 
of reflectors providing a total of 40 pairs 
of bulbs. The connections to the 
commutator were such that from 1 to 
8 bulbs could be connected in parallel 
and flashed simultaneously, and the 
speed of the commutator was set so that 
40 sets of bulbs were fired in sequence 
at such a rate that the maximum- 
intensity portion of the curves over- 
lapped to provide essentially continuous 
illumination. This arrangement oper- 
ated satisfactorily, although field ex- 
perience made it apparent that improved 
operation would result from moisture- 
proofing the electrical equipment and 
the use of a power source capable of 
supplying more instantaneous current 
than the 22^-v radio batteries. For 
field use this unit had the prime ad- 
vantage that the energy that supplied 
the illumination was stored chemically 
in the flashbulbs and required only 
triggering energy from a battery, while 
the motor driving the commutator could 
be supplied from the same generator 
that operated the high-speed cameras. 
With this equipment, it was possible to 
secure satisfactory pictures when the 
camera was operated at speeds of 
approximately 1000 frames/sec and 
the area to be photographed consisted 
of 400 sq ft or so. This area was lighted 
from a distance of approximately 120 
ft using 8 flashbulbs to the group; 
i.e., 320 flashbulbs were used to provide 
illumination for the two seconds needed 
to photograph the event. Two cameras 
were used to photograph the interior 
of these structures: the Bell & Howell 
70-S running at 128 frames/sec and the 
Fastax set at varying speeds between 
132 frames/sec and about 1200 frames/ 
sec. An observation of incidental in- 
terest is that the frame rate on the 
Bell & Howell camera was apparently 
sufficiently low for the negative to show 
no flicker in the light source, while the 



R. M. Blunt: Photography of Underground Explosions 



411 




412 



April 1953 Journal of the SMPTE Vol. 60 




R. M. Blunt: Photography of Underground Explosions 



413 




.2 -5 




414 



April 1953 Journal of the SMPTE Vol. 60 




Fig. 6. An illustration of the need for camera shelters. The 
equipment was not seriously damaged. 



pictures taken on the Fastax camera 
at the higher frame rate showed a 
definite pulsation in the intensity of the 
illumination. If it were a matter of 
any importance this could perhaps be 
overcome by operating the commutator 
at a different speed, to change the 
overlap on the peaks of the light output 
curves from the flashbulbs. The only 
serious damage sustained by the equip- 
ment used inside the structures is shown 
in Fig. 6. 

It is pleasant to be able to say that 



the Fastax cameras gave trouble-free 
operation in spite of rough handling 
and frequent exposure to adverse condi- 
tions of excessive dust and moisture. 
Toward the end of the program, auto- 
matic cutoff switches were installed 
which at first functioned beautifully, 
but after six weeks one of them stuck 
shut. Since the camera was timed to 
catch the event on the last 25 ft of the 
100-ft roll, most of the film on which 
the event was recorded was beaten to 
confetti during the lengthy over-run. 



R. M. Blunt: Photography of Underground Explosions 



415 






Therefore we returned to the use of a 
special control unit built specifically 
for this operation. These units con- 
tained heavy-duty relays designed to 
control much higher currents than the 
camera would draw, and pneumatically 
controlled time-delay relays that were 
set to provide about 10% more than the 
calculated running time before shutting 
off the camera. No failures were 
experienced with this equipment. 

It is our belief that the longer focal 
length lenses provided for the Fastax 
camera should be examined for excessive 
change in focal point when subjected to 
a temperature change of, say, 30 F. 
The pictures taken with the 10-in. lens 
were out of focus, and this may have 
been caused by the temperature rise 
that occurred between the setting up of 
the camera and its operation. Should 
an examination prove the lens to be 
temperature sensitive, precautions could 
be taken to minimize the effect through 
a redesign of the lens or, if this is not 
possible, to protect it from changes in 
temperature. 

In a test operation of this magnitude 
that involves a coordination of the high- 
speed cameras with a considerable 
variety of other instrumentation, we 
suggest that all circuits to the camera be 
centralized in a special control panel on 
which signal lamps would be arranged 
to indicate at all times the condition of 
each circuit. In addition, apparatus for 
recording the instant of arrival of the 
trigger signal at the panel and the 
subsequent behavior of the circuits 
should be provided. Although one may 
be accused of gilding the lily, a means of 
remotely controlling the Fastax speed 
and iris settings would have made it 
possible to secure better pictures in a 
number of instances. 

To the best of our knowledge, no 
really good equipment is available for 
reading the films from the high-speed 
cameras. We have used the Bell & 
Howell 16mm time-and-motion study 
projector for some years, but it is not 



well suited to our needs. The apparatus 
we would like to have would possess the 
following features: 

1. Ability to rotate the image 15 
about the optical axis. 

2. Ability to vary the magnification 
of the image continuously from 1 5 to 30 
times. 

3. Better framing, so that a minimum 
of adjustment is required to cause 
succeeding frames to coincide exactly 
on the screen, regardless of the direction 
in which the film has been moving. 

4. Forward and reverse viewing down 
to 4 frames/sec in addition to single-frame 
projection. 

5. The image should appear on a 
screen set into a horizontal working 
surface. The screen might well be set 
at a small angle, say 20, to facilitate 
viewing from a seated position. The 
screen should be replaceable with a piece 
of clear glass to permit projecting the 
image directly on graph paper so that 
one can obtain a direct tracing of the 
image. 

6. The illumination of the image 
should be high enough to permit tracing 
but adjustable by means of a diaphragm 
to a level comfortable for direct viewing 
on the ground glass. 

7. The entire apparatus should not 
cost over $1500. 

In order to provide the facilities 
required to process the films obtained on 
this program, a darkroom was set up 
in an 8 X 8 X 24 ft trailer. Cabinets 
and counters of the type employed in 
modern kitchens were installed and a 
large sink was provided. In addition 
to these fundamental items, a small 
refrigerator for the purpose of storing 
film, a 35mm contact printer, and a 
Houston machine were installed in the 
other end of the trailer. This amount of 
equipment did not leave a great deal of 
space for the personnel, and it is our 
feeling that a somewhat larger trailer 
would have permitted a more efficient 
installation, although this arrangement 
was satisfactory. In connection with 



416 



Aprtt 1953 Journal of the SMPTE Vol. 60 



processing in the field, one of the most 
serious problems is likely to be that of 
securing a satisfactory supply of water. 
It was our observation that the greatest 
difficulty, at least in processing black- 
and-white film, was likely to be the 
removal of suspended material. The 
hardness of the water did not appear 
to affect the processing. In future 
operations, we will consider it most 
important to provide means to remove 
suspended material and a source of pure 
water to make up the developing solu- 
tion. In this respect, water found in the 
Castledale, Utah, area gave us the 
greatest amount of trouble, since the 
material found in it was so finely divided 
that it passed through even a porous 
metallic type of filter and deposited a 
sludge on the photographic material 
that had to be wiped away. If the 
emulsion was permitted to dry, the 
material suspended in the water which 
had been deposited upon the surface of 
the film became almost a part of the 
emulsion and was impossible to remove. 
In the desert country in which we were 
operating, the installation of an effective 



air-conditioner in the trainer was most 
essential. The air-conditioning unit 
broke down for a few days and during 
this time it was not unusual to have the 
temperature in the trailer rise to 130 
to 140 F. It is, of course, almost im- 
possible to work under these conditions. 
In conclusion, I would like to point 
out that this program was carried on by 
several organizations besides the In- 
stitute of Industrial Research of the 
University of Denver. The Sacramento 
District, Corps of Engineers, had estab- 
lished the prime contract with Engineer- 
ing Research Associates who in turn had 
subcontracted portions of the work to the 
Armour Research Foundation, Rens- 
selaer Polytechnic Institute, and the 
Institute of Industrial Research. Other 
phases of the program were studied by 
Stanford Research Institute, Colorado 
School of Mines, and the U.S. Bureau 
of Mines. It is a pleasure for me to 
record here my indebtedness to Dr. W. 
Ray Jewell, of the University of Denver, 
for his assistance in solving many of the 
problems that arose in connection with 
this program. 



R. M. Blunt: Photography of Underground Explosions 



417 



Slides and Opaques for 
Television Film Chains, PH22.94 



A PROPOSED STANDARD on slides and 
opaques for television use is published 
on the following pages for three months 
of trial and criticism. All comments 
should be sent to Henry Kogel, SMPTE 
Staff Engineer, prior to July 1, 1953. If 
no adverse comments are received, the 
proposal will then be submitted to ASA 
Sectional Committee PH22 for further 
processing as an American Standard. 
The proposal was prepared by the Joint 
RTMA/SMPTE Television Film Equip- 
ment Committee. 

Work on the proposal was started in 
1950. At that time there existed an 
American Standard on Lantern Slides 
(Z38.7. 149-1 950) but it was concerned 
only with slides intended for direct 
viewing on a projection screen. The 
inflexible demand of the television system 
for a 3 by 4 aspect ratio forced all users 
to depart from some of the provisions of 
Z38.7.149. As a result, numerous local 
standards came into use but there was 
little agreement among the standards of 
different organizations or different areas. 

When the Committee first explored 
this subject, it was realized that a major 
service could be performed simply by 



securing widespread agreement on the 
exact dimensions of each of the slide 
sizes in common use. It also seemed 
that an additional service might be 
rendered if agreement could be secured 
on the use of fewer sizes. 

In an initial burst of optimism, the 
Committee prepared a proposal based on 
the use of just two sizes. This proposal 
was widely circulated in 1951 among the 
membership of the SMPTE, the RTMA, 
and the National Association of Radio 
and Television Broadcasters. Responses 
to this first effort indicated that everyone 
felt it advisable to accept as standard 
no more than two sizes. However, 
there was little or no agreement on the 
two sizes that should be selected for 
standardization. Altogether nine sizes 
were suggested for standardization. The 
Committee was able to cut this number 
back to four by insisting that widespread 
usage at this time was a necessary condi- 
tion for adoption of any size. 

At no time has there been any serious 
objection to the exact dimensions in- 
cluded in the Proposed Standard. Users 
seem to be content with any number so 
long as everyone accepts it. F. N. 
Gillette. 



418 



April 1953 Journal of the SMPTE Vol. 60 



Proposed American Standard 

Slides and Opaques 
for Television Film Chains 

(Second Draft) 



PH22.94 



Pag. 1 Of 2 page, 



1. Scope 

1.1 This proposal is intended to supplement 
American Standard Z38.7. 19-1 950, not re- 
place it. The television system imposes special 
requirements that did not enter into the prep- 
aration of 138.7. 19-1 950. 

1.2 The proposal applies only to slides and 
opaques intended for transmission in the 
standard fashion via a film chain. For other 
applications, such as background projection, 
the usual television requirements may not 
apply. 



2. Standard 

2.1 Nominal Size. Only the four nominal 
sizes listed in column 1 of the table shall be 
considered standard for use in television film 
chains. 



2.2 Overall Dimensions. The overall di- 
mensions for any nominal size shall comply 
with the dimensions tabulated in column 2. 
(See Note 1.) 

2.3 Dimensions of Transmitted Picture. 
The portion of the slide or opaque intended 
for transmission shall lie within a centrally 
located rectangle having the dimensions 
shown in column 3. (See Note 2.) 

2.4 Dimensions of Picture Background. 
The background of the slide or opaque shall 
extend without interruption over a centrally 
located rectangle having the dimensions 
shown in column 4. (See Note 2.) 

2.5 Centering Tolerance. The center of 
the transmitted picture rectangle and the cen- 
ter of the background rectangle shall both Re 
within a circle having as its center the center 
of the slide and as its radius the dimension 
tabulated in column 5. 



Note 1. The overall dimensions are in accord with 
American Standard Z38.7.19 1950 insofar as it is 
relevant. The thickness of opaques is not covered by 
Z38.7. 19-1 950. This quantity is here specified as 
1/32 inch on the assumption that opaques will con- 
sist of double-weight photographic power without 
additional support or backing. 

Note 2. The dimensions shown for the transmitted 
picture are those which will be scanned by a perfectly 
adjusted film chain. To allow for some misadjustment 
of the film chain and an additional misadjustment 



in the home receiver, it is recommended that all essen- 
tial information be contained in a centrally located 
area appreciably smaller than that specified in 
column 3. 

Note 3. In the case of slides, the background rec- 
tangle should be defined by an opaque mask to limit 
the stray light entering the film chain. The dimensions 
specified in column 4 permit the use of masks which 
comply with Z38.7.19 1950. For opaques, masking is 
generally provided by the projection equipment. 



NOT APPROVED 



April 1953 Journal of the SMPTE Vol. 60 



419 



Page 2 of